UNIT 1A
Overview
The idea of data warehousing came to the late 1980's when IBM researchers Barry Devlin and Paul Murphy established the "Business Data Warehouse."
Data Warehouse is a relational database management system (RDBMS) construct to meet the requirement of transaction processing systems. It can be described as any centralized data repository which can be queried for business benefits. It is a database that stores information oriented to satisfy decision-making requests. It is a group of decision support technologies, targets to enabling the knowledge worker (executive, manager, and analyst) to make superior and higher decisions. So, Data Warehousing support architectures and tool for business executives to systematically organize, understand and use their information to make strategic decisions.
Data Warehouse environment contains an extraction, transportation, and loading (ETL) solution, an online analytical processing (OLAP) engine, customer analysis tools, and other applications that handle the process of gathering information and delivering it to business users.
What is a Data Warehouse?
A Data Warehouse (DW) is a relational database that is designed for query and analysis rather than transaction processing. It includes historical data derived from transaction data from single and multiple sources.
A Data Warehouse provides integrated, enterprise-wide, historical data and focuses on providing support for decision-makers for data modelling and analysis.
Characteristics of Data Warehouse
"Data Warehouse is a subject-oriented, integrated, and time-variant store of information in support of management's decisions."
Subject-Oriented
A data warehouse target on the modeling and analysis of data for decision-makers. Therefore, data warehouses typically provide a concise and straightforward view around a particular subject, such as customer, product, or sales, instead of the global organization's ongoing operations. This is done by excluding data that are not useful concerning the subject and including all data needed by the users to understand the subject.
Integrated
A data warehouse integrates various heterogeneous data sources like RDBMS, flat files, and online transaction records. It requires performing data cleaning and integration during data warehousing to ensure consistency in naming conventions, attributes types, etc., among different data sources.
Time-Variant
Historical information is kept in a data warehouse. For example, one can retrieve files from 3 months, 6 months, 12 months, or even previous data from a data warehouse. These variations with a transactions system, where often only the most current file is kept.
Non-Volatile
The data warehouse is a physically separate data storage, which is transformed from the source operational RDBMS. The operational updates of data do not occur in the data warehouse, i.e., update, insert, and delete operations are not performed. It usually requires only two procedures in data accessing: Initial loading of data and access to data. Therefore, the DW does not require transaction processing, recovery, and concurrency capabilities, which allows for substantial speedup of data retrieval. Non-Volatile defines that once entered into the warehouse, and data should not change.
Goals of Data Warehousing
- To help reporting as well as analysis
- Maintain the organization's historical information
- Be the foundation for decision making.
Need for Data Warehouse
Data Warehouse is needed for the following reasons:
- Business User: Business users require a data warehouse to view summarized data from the past. Since these people are non-technical, the data may be presented to them in an elementary form.
- Store historical data: Data Warehouse is required to store the time variable data from the past. This input is made to be used for various purposes.
- Make strategic decisions: Some strategies may be depending upon the data in the data warehouse. So, data warehouse contributes to making strategic decisions.
- For data consistency and quality: Bringing the data from different sources at a commonplace, the user can effectively undertake to bring the uniformity and consistency in data.
- High response time: Data warehouse has to be ready for somewhat unexpected loads and types of queries, which demands a significant degree of flexibility and quick response time.
Benefits of Data Warehouse
- Understand business trends and make better forecasting decisions.
- Data Warehouses are designed to perform well enormous amounts of data.
- The structure of data warehouses is more accessible for end-users to navigate, understand, and query.
- Queries that would be complex in many normalized databases could be easier to build and maintain in data warehouses.
- Data warehousing is an efficient method to manage demand for lots of information from lots of users.
- Data warehousing provide the capabilities to analyze a large amount of historical data.
Components or Building Blocks of Data Warehouse
1. Source Data Component
Source data coming into the data warehouses may be grouped into four broad categories:
Production Data: This type of data comes from the different operating systems of the enterprise. Based on the data requirements in the data warehouse, we choose segments of the data from the various operational modes.
Internal Data: In each organization, the client keeps their "private" spreadsheets, reports, customer profiles, and sometimes even department databases. This is the internal data, part of which could be useful in a data warehouse.
Archived Data: Operational systems are mainly intended to run the current business. In every operational system, we periodically take the old data and store it in archived files.
External Data: Most executives depend on information from external sources for a large percentage of the information they use. They use statistics associating to their industry produced by the external department.
2. Data Staging Component
After we have been extracted data from various operational systems and external sources, we have to prepare the files for storing in the data warehouse. The extracted data coming from several different sources need to be changed, converted, and made ready in a format that is relevant to be saved for querying and analysis.
We will now discuss the three primary functions that take place in the staging area.
1) Data Extraction: This method has to deal with numerous data sources. We have to employ the appropriate techniques for each data source.
2) Data Transformation: As we know, data for a data warehouse comes from many different sources. If data extraction for a data warehouse posture big challenges, data transformation presents even significant challenges. We perform several individual tasks as part of data transformation.
First, we clean the data extracted from each source. Cleaning may be the correction of misspellings or may deal with providing default values for missing data elements, or elimination of duplicates when we bring in the same data from various source systems.
Standardization of data components forms a large part of data transformation. Data transformation contains many forms of combining pieces of data from different sources. We combine data from single source record or related data parts from many source records.
On the other hand, data transformation also contains purging source data that is not useful and separating outsource records into new combinations. Sorting and merging of data take place on a large scale in the data staging area. When the data transformation function ends, we have a collection of integrated data that is cleaned, standardized, and summarized.
3) Data Loading: Two distinct categories of tasks form data loading functions. When we complete the structure and construction of the data warehouse and go live for the first time, we do the initial loading of the information into the data warehouse storage. The initial load moves high volumes of data using up a substantial amount of time.
3. Data Storage Components
Data storage for the data warehousing is a split repository. The data repositories for the operational systems generally include only the current data. Also, these data repositories include the data structured in highly normalized for fast and efficient processing.
4. Information Delivery Component
The information delivery element is used to enable the process of subscribing for data warehouse files and having it transferred to one or more destinations according to some customer-specified scheduling algorithm.
5. Metadata Component
Metadata in a data warehouse is equal to the data dictionary or the data catalogue in a database management system. In the data dictionary, we keep the data about the logical data structures, the data about the records and addresses, the information about the indexes, and so on.
6. Data Marts
It includes a subset of corporate-wide data that is of value to a specific group of users. The scope is confined to particular selected subjects. Data in a data warehouse should be a fairly current, but not mainly up to the minute, although development in the data warehouse industry has made standard and incremental data dumps more achievable. Data marts are lower than data warehouses and usually contain organization. The current trends in data warehousing are to developed a data warehouse with several smaller related data marts for particular kinds of queries and reports.
7. Management and Control Component
The management and control elements coordinate the services and functions within the data warehouse. These components control the data transformation and the data transfer into the data warehouse storage. On the other hand, it moderates the data delivery to the clients. Its work with the database management systems and authorizes data to be correctly saved in the repositories. It monitors the movement of information into the staging method and from there into the data warehouses storage itself.
Why we need a separate Data Warehouse?
Data Warehouse queries are complex because they involve the computation of large groups of data at summarized levels.
It may require the use of distinctive data organization, access, and implementation method based on multidimensional views.
Performing OLAP queries in operational database degrade the performance of functional tasks.
Data Warehouse is used for analysis and decision making in which extensive database is required, including historical data, which operational database does not typically maintain.
The separation of an operational database from data warehouses is based on the different structures and uses of data in these systems.
Because the two systems provide different functionalities and require different kinds of data, it is necessary to maintain separate databases.
Difference between Database and Data Warehouse
Database | Data Warehouse |
1. It is used for Online Transactional Processing (OLTP) but can be used for other objectives such as Data Warehousing. This records the data from the clients for history. | 1. It is used for Online Analytical Processing (OLAP). This reads the historical information for the customers for business decisions. |
2. The tables and joins are complicated since they are normalized for RDBMS. This is done to reduce redundant files and to save storage space. | 2. The tables and joins are accessible since they are de-normalized. This is done to minimize the response time for analytical queries. |
3. Data is dynamic | 3. Data is largely static |
4. Entity: Relational modeling procedures are used for RDBMS database design. | 4. Data: Modeling approach are used for the Data Warehouse design. |
5. Optimized for write operations. | 5. Optimized for read operations. |
6. Performance is low for analysis queries. | 6. High performance for analytical queries. |
7. The database is the place where the data is taken as a base and managed to get available fast and efficient access. | 7. Data Warehouse is the place where the application data is handled for analysis and reporting objectives. |
Building a Data Warehouse
Some steps that are needed for building any data warehouse are as following below:
- To extract the data (transnational) from different data sources: For building a data warehouse, a data is extracted from various data sources and that data is stored in central storage area. For extraction of the data Microsoft has come up with an excellent tool. When you purchase Microsoft SQL Server, then this tool will be available at free of cost.
- To transform the transactional data: There are various DBMS where many of the companies stores their data. Some of them are: MS Access, MS SQL Server, Oracle, Sybase etc. Also these companies saves the data in spreadsheets, flat files, mail systems etc. Relating a data from all these sources is done while building a data warehouse.
- To load the data (transformed) into the dimensional database: After building a dimensional model, the data is loaded in the dimensional database. This process combines the several columns together or it may split one field into the several columns. There are two stages at which transformation of the data can be performed and they are: while loading the data into the dimensional model or while data extraction from their origins.
- To purchase a front-end reporting tool: There are top notch analytical tools are available in the market. These tools are provided by the several major vendors. A cost effective tool and Data Analyzer is released by the Microsoft on its own.
Data Warehouse Architecture
A data warehouse architecture is a method of defining the overall architecture of data communication processing and presentation that exist for end-clients computing within the enterprise. Each data warehouse is different, but all are characterized by standard vital components.
Production applications such as payroll accounts payable product purchasing and inventory control are designed for online transaction processing (OLTP). Such applications gather detailed data from day to day operations.
Data Warehouse applications are designed to support the user ad-hoc data requirements, an activity recently dubbed online analytical processing (OLAP). These include applications such as forecasting, profiling, summary reporting, and trend analysis.
Production databases are updated continuously by either by hand or via OLTP applications. In contrast, a warehouse database is updated from operational systems periodically, usually during off-hours. As OLTP data accumulates in production databases, it is regularly extracted, filtered, and then loaded into a dedicated warehouse server that is accessible to users. As the warehouse is populated, it must be restructured tables de-normalized, data cleansed of errors and redundancies and new fields and keys added to reflect the needs to the user for sorting, combining, and summarizing data.
Data warehouses and their architectures very depending upon the elements of an organization's situation.
Three common architectures are:
- Data Warehouse Architecture: Basic
- Data Warehouse Architecture: With Staging Area
- Data Warehouse Architecture: With Staging Area and Data Marts
Data Warehouse Architecture: Basic
Operational System
An operational system is a method used in data warehousing to refer to a system that is used to process the day-to-day transactions of an organization.
Flat Files
A Flat file system is a system of files in which transactional data is stored, and every file in the system must have a different name.
Meta Data
A set of data that defines and gives information about other data.
Meta Data used in Data Warehouse for a variety of purpose, including:
Meta Data summarizes necessary information about data, which can make finding and work with particular instances of data more accessible. For example, author, data build, and data changed, and file size are examples of very basic document metadata.
Metadata is used to direct a query to the most appropriate data source.
Lightly and highly summarized data
The area of the data warehouse saves all the predefined lightly and highly summarized (aggregated) data generated by the warehouse manager.
The goals of the summarized information are to speed up query performance. The summarized record is updated continuously as new information is loaded into the warehouse.
End-User access Tools
The principal purpose of a data warehouse is to provide information to the business managers for strategic decision-making. These customers interact with the warehouse using end-client access tools.
The examples of some of the end-user access tools can be:
- Reporting and Query Tools
- Application Development Tools
- Executive Information Systems Tools
- Online Analytical Processing Tools
- Data Mining Tools
Data Warehouse Architecture: With Staging Area
We must clean and process your operational information before put it into the warehouse.
We can do this programmatically, although data warehouses uses a staging area (A place where data is processed before entering the warehouse).
A staging area simplifies data cleansing and consolidation for operational method coming from multiple source systems, especially for enterprise data warehouses where all relevant data of an enterprise is consolidated.
Data Warehouse Staging Area is a temporary location where a record from source systems is copied.
Data Warehouse Architecture: With Staging Area and Data Marts
We may want to customize our warehouse's architecture for multiple groups within our organization.
We can do this by adding data marts. A data mart is a segment of a data warehouses that can provided information for reporting and analysis on a section, unit, department or operation in the company, e.g., sales, payroll, production, etc.
The figure illustrates an example where purchasing, sales, and stocks are separated. In this example, a financial analyst wants to analyze historical data for purchases and sales or mine historical information to make predictions about customer behavior.
Properties of Data Warehouse Architectures
The following architecture properties are necessary for a data warehouse system:
1. Separation: Analytical and transactional processing should be keep apart as much as possible.
2. Scalability: Hardware and software architectures should be simple to upgrade the data volume, which has to be managed and processed, and the number of user's requirements, which have to be met, progressively increase.
3. Extensibility: The architecture should be able to perform new operations and technologies without redesigning the whole system.
4. Security: Monitoring accesses are necessary because of the strategic data stored in the data warehouses.
5. Administrability: Data Warehouse management should not be complicated.
Types of Data Warehouse Architectures
1. Single-Tier Architecture
Single-Tier architecture is not periodically used in practice. Its purpose is to minimize the amount of data stored to reach this goal; it removes data redundancies.
The figure shows the only layer physically available is the source layer. In this method, data warehouses are virtual. This means that the data warehouse is implemented as a multidimensional view of operational data created by specific middleware, or an intermediate processing layer.
The vulnerability of this architecture lies in its failure to meet the requirement for separation between analytical and transactional processing. Analysis queries are agreed to operational data after the middleware interprets them. In this way, queries affect transactional workloads.
2. Two-Tier Architecture
The requirement for separation plays an essential role in defining the two-tier architecture for a data warehouse system, as shown in fig:
Although it is typically called two-layer architecture to highlight a separation between physically available sources and data warehouses, in fact, consists of four subsequent data flow stages:
- Source layer: A data warehouse system uses a heterogeneous source of data. That data is stored initially to corporate relational databases or legacy databases, or it may come from an information system outside the corporate walls.
- Data Staging: The data stored to the source should be extracted, cleansed to remove inconsistencies and fill gaps, and integrated to merge heterogeneous sources into one standard schema. The so-named Extraction, Transformation, and Loading Tools (ETL) can combine heterogeneous schemata, extract, transform, cleanse, validate, filter, and load source data into a data warehouse.
- Data Warehouse layer: Information is saved to one logically centralized individual repository: a data warehouse. The data warehouses can be directly accessed, but it can also be used as a source for creating data marts, which partially replicate data warehouse contents and are designed for specific enterprise departments. Meta-data repositories store information on sources, access procedures, data staging, users, data mart schema, and so on.
- Analysis: In this layer, integrated data is efficiently, and flexible accessed to issue reports, dynamically analyze information, and simulate hypothetical business scenarios. It should feature aggregate information navigators, complex query optimizers, and customer-friendly GUIs.
3. Three-Tier Architecture
The three-tier architecture consists of the source layer (containing multiple source system), the reconciled layer and the data warehouse layer (containing both data warehouses and data marts). The reconciled layer sits between the source data and data warehouse.
The main advantage of the reconciled layer is that it creates a standard reference data model for a whole enterprise. At the same time, it separates the problems of source data extraction and integration from those of data warehouse population. In some cases, the reconciled layer is also directly used to accomplish better some operational tasks, such as producing daily reports that cannot be satisfactorily prepared using the corporate applications or generating data flows to feed external processes periodically to benefit from cleaning and integration.
This architecture is especially useful for the extensive, enterprise-wide systems. A disadvantage of this structure is the extra file storage space used through the extra redundant reconciled layer. It also makes the analytical tools a little further away from being real-time.
Three-Tier Data Warehouse Architecture
Data Warehouses usually have a three-level (tier) architecture that includes:
- Bottom Tier (Data Warehouse Server)
- Middle Tier (OLAP Server)
- Top Tier (Front end Tools).
A bottom-tier that consists of the Data Warehouse server, which is almost always an RDBMS. It may include several specialized data marts and a metadata repository.
Data from operational databases and external sources (such as user profile data provided by external consultants) are extracted using application program interfaces called a gateway. A gateway is provided by the underlying DBMS and allows customer programs to generate SQL code to be executed at a server.
Examples of gateways contain ODBC (Open Database Connection) and OLE-DB (Open-Linking and Embedding for Databases), by Microsoft, and JDBC (Java Database Connection).
A middle-tier which consists of an OLAP server for fast querying of the data warehouse.
The OLAP server is implemented using either
(1) A Relational OLAP (ROLAP) model, i.e., an extended relational DBMS that maps functions on multidimensional data to standard relational operations.
(2) A Multidimensional OLAP (MOLAP) model, i.e., a particular purpose server that directly implements multidimensional information and operations.
A top-tier that contains front-end tools for displaying results provided by OLAP, as well as additional tools for data mining of the OLAP-generated data.
The overall Data Warehouse Architecture is shown in fig:
The metadata repository stores information that defines DW objects. It includes the following parameters and information for the middle and the top-tier applications:
- A description of the DW structure, including the warehouse schema, dimension, hierarchies, data mart locations, and contents, etc.
- Operational metadata, which usually describes the currency level of the stored data, i.e., active, archived or purged, and warehouse monitoring information, i.e., usage statistics, error reports, audit, etc.
- System performance data, which includes indices, used to improve data access and retrieval performance.
- Information about the mapping from operational databases, which provides source RDBMSs and their contents, cleaning and transformation rules, etc.
- Summarization algorithms, predefined queries, and reports business data, which include business terms and definitions, ownership information, etc.
Types of Database Parallelism
Parallelism is used to support speedup, where queries are executed faster because more resources, such as processors and disks, are provided. Parallelism is also used to provide scale-up, where increasing workloads are managed without increase response-time, via an increase in the degree of parallelism.
Different architectures for parallel database systems are shared-memory, shared-disk, shared-nothing, and hierarchical structures.
(a) Horizontal Parallelism: It means that the database is partitioned across multiple disks, and parallel processing occurs within a specific task (i.e., table scan) that is performed concurrently on different processors against different sets of data.
(b) Vertical Parallelism: It occurs among various tasks. All component query operations (i.e., scan, join, and sort) are executed in parallel in a pipelined fashion. In other words, an output from one function (e.g., join) as soon as records become available.
Intraquery Parallelism
Intraquery parallelism defines the execution of a single query in parallel on multiple processors and disks. Using intraquery parallelism is essential for speeding up long-running queries.
Interquery parallelism does not help in this function since each query is run sequentially.
To improve the situation, many DBMS vendors developed versions of their products that utilized intraquery parallelism.
This application of parallelism decomposes the serial SQL, query into lower-level operations such as scan, join, sort, and aggregation.
These lower-level operations are executed concurrently, in parallel.
Interquery Parallelism
In interquery parallelism, different queries or transaction execute in parallel with one another.
This form of parallelism can increase transactions throughput. The response times of individual transactions are not faster than they would be if the transactions were run in isolation.
Thus, the primary use of interquery parallelism is to scale up a transaction processing system to support a more significant number of transactions per second.
Database vendors started to take advantage of parallel hardware architectures by implementing multiserver and multithreaded systems designed to handle a large number of client requests efficiently.
This approach naturally resulted in interquery parallelism, in which different server threads (or processes) handle multiple requests at the same time.
Interquery parallelism has been successfully implemented on SMP systems, where it increased the throughput and allowed the support of more concurrent users.
Design of Parallel Databases
A parallel DBMS is a DBMS that runs across multiple processors or CPUs and is mainly designed to execute query operations in parallel, wherever possible. The parallel DBMS link a number of smaller machines to achieve the same throughput as expected from a single large machine.
In Parallel Databases, mainly there are three architectural designs for parallel DBMS. They are as follows:
- Shared Memory Architecture
- Shared Disk Architecture
- Shared Nothing Architecture
Let’s discuss them one by one:
1. Shared Memory Architecture- In Shared Memory Architecture, there are multiple CPUs that are attached to an interconnection network. They are able to share a single or global main memory and common disk arrays. It is to be noted that, In this architecture, a single copy of a multi-threaded operating system and multithreaded DBMS can support these multiple CPUs. Also, the shared memory is a solid coupled architecture in which multiple CPUs share their memory. It is also known as Symmetric multiprocessing (SMP). This architecture has a very wide range which starts from personal workstations that support a few microprocessors in parallel via RISC.
Advantages :
- It has high-speed data access for a limited number of processors.
- The communication is efficient.
Disadvantages :
- It cannot use beyond 80 or 100 CPUs in parallel.
- The bus or the interconnection network gets block due to the increment of the large number of CPUs.
2. Shared Disk Architectures :
In Shared Disk Architecture, various CPUs are attached to an interconnection network. In this, each CPU has its own memory and all of them have access to the same disk. Also, note that here the memory is not shared among CPUs therefore each node has its own copy of the operating system and DBMS. Shared disk architecture is a loosely coupled architecture optimized for applications that are inherently centralized. They are also known as clusters.
Advantages :
- The interconnection network is no longer a bottleneck each CPU has its own memory.
- Load-balancing is easier in shared disk architecture.
- There is better fault tolerance.
Disadvantages :
- If the number of CPUs increases, the problems of interference and memory contentions also increase.
- There’s also exists a scalability problem.
3. Shared Nothing Architecture :
Shared Nothing Architecture is multiple processor architecture in which each processor has its own memory and disk storage. In this, multiple CPUs are attached to an interconnection network through a node. Also, note that no two CPUs can access the same disk area. In this architecture, no sharing of memory or disk resources is done. It is also known as Massively parallel processing (MPP).
Advantages :
- It has better scalability as no sharing of resources is done
- Multiple CPUs can be added
Disadvantages:
- The cost of communications is higher as it involves sending of data and software interaction at both ends
- The cost of non-local disk access is higher than the cost of shared disk architectures.
Note that this technology is typically used for very large databases that have the size of 1012 bytes or TB or for the system that has the process of thousands of transactions per second.
4, Hierarchical Architecture :
This architecture is a combination of shared disk, shared memory and shared nothing architectures. This architecture is scalable due to availability of more memory and many processor. But is costly to other architecture.
MultiDimensional Data Model
The multi-Dimensional Data Model is a method which is used for ordering data in the database along with good arrangement and assembling of the contents in the database.
The Multi Dimensional Data Model allows customers to interrogate analytical questions associated with market or business trends, unlike relational databases which allow customers to access data in the form of queries. They allow users to rapidly receive answers to the requests which they made by creating and examining the data comparatively fast.
OLAP (online analytical processing) and data warehousing uses multi dimensional databases. It is used to show multiple dimensions of the data to users.
It represents data in the form of data cubes. Data cubes allow to model and view the data from many dimensions and perspectives. It is defined by dimensions and facts and is represented by a fact table. Facts are numerical measures and fact tables contain measures of the related dimensional tables or names of the facts.
Working on a Multidimensional Data Model
On the basis of the pre-decided steps, the Multidimensional Data Model works.
The following stages should be followed by every project for building a Multi Dimensional Data Model :
Stage 1 : Assembling data from the client : In first stage, a Multi Dimensional Data Model collects correct data from the client. Mostly, software professionals provide simplicity to the client about the range of data which can be gained with the selected technology and collect the complete data in detail.
Stage 2 : Grouping different segments of the system : In the second stage, the Multi Dimensional Data Model recognizes and classifies all the data to the respective section they belong to and also builds it problem-free to apply step by step.
Stage 3 : Noticing the different proportions : In the third stage, it is the basis on which the design of the system is based. In this stage, the main factors are recognized according to the user’s point of view. These factors are also known as “Dimensions”.
Stage 4 : Preparing the actual-time factors and their respective qualities : In the fourth stage, the factors which are recognized in the previous step are used further for identifying the related qualities. These qualities are also known as “attributes” in the database.
Stage 5 : Finding the actuality of factors which are listed previously and their qualities : In the fifth stage, A Multi Dimensional Data Model separates and differentiates the actuality from the factors which are collected by it. These actually play a significant role in the arrangement of a Multi Dimensional Data Model.
Stage 6 : Building the Schema to place the data, with respect to the information collected from the steps above : In the sixth stage, on the basis of the data which was collected previously, a Schema is built.
For Example :
- Let us take the example of a firm. The revenue cost of a firm can be recognized on the basis of different factors such as geographical location of firm’s workplace, products of the firm, advertisements done, time utilized to flourish a product, etc.
- Let us take the example of the data of a factory which sells products per quarter in Bangalore. The data is represented in the table given below :
In the above given presentation, the factory’s sales for Bangalore are, for the time dimension, which is organized into quarters and the dimension of items, which is sorted according to the kind of item which is sold. The facts here are represented in rupees (in thousands).
Now, if we desire to view the data of the sales in a three-dimensional table, then it is represented in the diagram given below. Here the data of the sales is represented as a two dimensional table. Let us consider the data according to item, time and location (like Kolkata, Delhi, Mumbai). Here is the table :
This data can be represented in the form of three dimensions conceptually, which is shown in the image below :
Features of multidimensional data models:
Measures: Measures are numerical data that can be analyzed and compared, such as sales or revenue. They are typically stored in fact tables in a multidimensional data model.
Dimensions: Dimensions are attributes that describe the measures, such as time, location, or product. They are typically stored in dimension tables in a multidimensional data model.
Cubes: Cubes are structures that represent the multidimensional relationships between measures and dimensions in a data model. They provide a fast and efficient way to retrieve and analyze data.
Aggregation: Aggregation is the process of summarizing data across dimensions and levels of detail. This is a key feature of multidimensional data models, as it enables users to quickly analyze data at different levels of granularity.
Drill-down and roll-up: Drill-down is the process of moving from a higher-level summary of data to a lower level of detail, while roll-up is the opposite process of moving from a lower-level detail to a higher-level summary. These features enable users to explore data in greater detail and gain insights into the underlying patterns.
Hierarchies: Hierarchies are a way of organizing dimensions into levels of detail. For example, a time dimension might be organized into years, quarters, months, and days. Hierarchies provide a way to navigate the data and perform drill-down and roll-up operations.
OLAP (Online Analytical Processing): OLAP is a type of multidimensional data model that supports fast and efficient querying of large datasets. OLAP systems are designed to handle complex queries and provide fast response times.
Advantages of Multi Dimensional Data Model
The following are the advantages of a multi-dimensional data model :
- A multi-dimensional data model is easy to handle.
- It is easy to maintain.
- Its performance is better than that of normal databases (e.g. relational databases).
- The representation of data is better than traditional databases. That is because the multi-dimensional databases are multi-viewed and carry different types of factors.
- It is workable on complex systems and applications, contrary to the simple one-dimensional database systems.
- The compatibility in this type of database is an upliftment for projects having lower bandwidth for maintenance staff.
Disadvantages of Multi Dimensional Data Model
The following are the disadvantages of a Multi Dimensional Data Model :
- The multi-dimensional Data Model is slightly complicated in nature and it requires professionals to recognize and examine the data in the database.
- During the work of a Multi-Dimensional Data Model, when the system caches, there is a great effect on the working of the system.
- It is complicated in nature due to which the databases are generally dynamic in design.
- The path to achieving the end product is complicated most of the time.
- As the Multi Dimensional Data Model has complicated systems, databases have a large number of databases due to which the system is very insecure when there is a security break.
Data Warehousing - Schemas
Schema is a logical description of the entire database. It includes the name and description of records of all record types including all associated data-items and aggregates. Much like a database, a data warehouse also requires to maintain a schema. A database uses relational model, while a data warehouse uses Star, Snowflake, and Fact Constellation schema. In this chapter, we will discuss the schemas used in a data warehouse.
Star Schema
- Each dimension in a star schema is represented with only one-dimension table.
- This dimension table contains the set of attributes.
- The following diagram shows the sales data of a company with respect to the four dimensions, namely time, item, branch, and location.
- There is a fact table at the center. It contains the keys to each of four dimensions.
- The fact table also contains the attributes, namely dollars sold and units sold.
Note − Each dimension has only one dimension table and each table holds a set of attributes. For example, the location dimension table contains the attribute set {location_key, street, city, province_or_state,country}. This constraint may cause data redundancy. For example, "Vancouver" and "Victoria" both the cities are in the Canadian province of British Columbia. The entries for such cities may cause data redundancy along the attributes province_or_state and country.
Snowflake Schema
- Some dimension tables in the Snowflake schema are normalized.
- The normalization splits up the data into additional tables.
- Unlike Star schema, the dimensions table in a snowflake schema are normalized. For example, the item dimension table in star schema is normalized and split into two dimension tables, namely item and supplier table.
- Now the item dimension table contains the attributes item_key, item_name, type, brand, and supplier-key.
- The supplier key is linked to the supplier dimension table. The supplier dimension table contains the attributes supplier_key and supplier_type.
Note − Due to normalization in the Snowflake schema, the redundancy is reduced and therefore, it becomes easy to maintain and the save storage space.
Fact Constellation Schema
- A fact constellation has multiple fact tables. It is also known as galaxy schema.
- The following diagram shows two fact tables, namely sales and shipping.
- The sales fact table is same as that in the star schema.
- The shipping fact table has the five dimensions, namely item_key, time_key, shipper_key, from_location, to_location.
- The shipping fact table also contains two measures, namely dollars sold and units sold.
- It is also possible to share dimension tables between fact tables. For example, time, item, and location dimension tables are shared between the sales and shipping fact table.
Schema Definition
Multidimensional schema is defined using Data Mining Query Language (DMQL). The two primitives, cube definition and dimension definition, can be used for defining the data warehouses and data marts.
What is a Concept Hierarchies?
A concept hierarchy represents a series of mappings from a set of low-level concepts to larger-level, more general concepts. Concept hierarchy organizes information or concepts in a hierarchical structure or a specific partial order, which are used for defining knowledge in brief, high-level methods, and creating possible mining knowledge at several levels of abstraction.
A conceptual hierarchy includes a set of nodes organized in a tree, where the nodes define values of an attribute known as concepts. A specific node, “ANY”, is constrained for the root of the tree. A number is created to the level of each node in a conceptual hierarchy. The level of the root node is one. The level of a non-root node is one more the level of its parent level number.
Because values are defined by nodes, the levels of nodes can also be used to describe the levels of values. Concept hierarchy enables raw information to be managed at a higher and more generalized level of abstraction. There are several types of concept hierarchies which are as follows −
Schema Hierarchy − Schema hierarchy represents the total or partial order between attributes in the database. It can define existing semantic relationships between attributes. In a database, more than one schema hierarchy can be generated by using multiple sequences and grouping of attributes.
Set-Grouping Hierarchy − A set-grouping hierarchy constructs values for a given attribute or dimension into groups or constant range values. It is also known as instance hierarchy because the partial series of the hierarchy is represented on the set of instances or values of an attribute. These hierarchies have more functional sense and are so approved than other hierarchies.
Operation-Derived Hierarchy − Operation-derived hierarchy is represented by a set of operations on the data. These operations are defined by users, professionals, or the data mining system. These hierarchies are usually represented for mathematical attributes. Such operations can be as easy as range value comparison, as difficult as a data clustering and data distribution analysis algorithm.
Rule-based Hierarchy − In a rule-based hierarchy either a whole concept hierarchy or an allocation of it is represented by a set of rules and is computed dynamically based on the current information and rule definition. A lattice-like architecture is used for graphically defining this type of hierarchy, in which each child-parent route is connected with a generalization rule.
The static and dynamic generation of concept hierarchy is based on data sets. In this context, the generation of a concept hierarchy depends on a static or dynamic data set is known as the static or dynamic generation of concept hierarchy.
How are concept hierarchies useful in OLAP?
In the multidimensional model, data are arranged into several dimensions, and each dimension includes several levels of abstraction represented by concept hierarchies. This organization supports users with the adaptability to view records from different perspectives.
Several OLAP data cube operations continue to materialize these multiple views, enabling interactive querying and analysis of the data at hand. Therefore, OLAP supports a convenient environment for interactive data analysis.
Five basic OLAP commands are used to implement data retrieval from a Data warehouse are as follows −
ROLL UP Command − The ROLL UP enables the user to summarise information into a higher general level in the hierarchy. The roll-up operation shown aggregates the records by ascending the location hierarchy from the level of the city to the level of the country. In other terms, instead of grouping the data by city, the resulting cube groups the data by country.
When roll-up is executed by dimension reduction, one or more dimensions are deleted from the given cube. For instance, consider a sales data cube including only the two dimensions location and time. Roll-up can be implemented by removing, say, the time dimension, resulting in an aggregation of the total sales by location, instead of by location and by time.
Drill-down** − Drill-down is the reverse of roll-up. It operates from less detailed information to more detailed information. Drill-down can be completed by either stepping down a concept hierarchy for a dimension or presenting more dimensions. Drill-down appears by descending the time hierarchy from the level of the quarter to the more detailed level of the month. The resulting data cube analysis the total sales per month instead of summarizing them by quarter.
Slice and dice − The slice operation implemented a selection on one dimension of the given cube, resulting in a subcube. The dice operation represents a subcube by implementing a selection on two or more dimensions.
Pivot − Pivot is also known as rotate. It is a visualization operation that rotates the data axes in view to support an alternative performance of the data.
Other OLAP operations − Some OLAP systems provide more drilling operations. For instance, drill-across implements queries containing (i.e., across) more than one fact table. The drill-through operation uses relational SQL services to drill through the bottom level of a data cube down to its back-end relational tables.
Several OLAP operations can involve ranking the top N or bottom N items in lists, and calculating moving averages, growth values, and interests, internal values of return, depreciation, currency conversions, and statistical services.
Characteristics of OLAP
In the FASMI characteristics of OLAP methods, the term derived from the first letters of the characteristics are:
Fast
It defines which the system targeted to deliver the most feedback to the client within about five seconds, with the elementary analysis taking no more than one second and very few taking more than 20 seconds.
Analysis
It defines which the method can cope with any business logic and statistical analysis that is relevant for the function and the user, keep it easy enough for the target client. Although some preprogramming may be needed we do not think it acceptable if all application definitions have to be allow the user to define new Adhoc calculations as part of the analysis and to document on the data in any desired method, without having to program so we excludes products (like Oracle Discoverer) that do not allow the user to define new Adhoc calculation as part of the analysis and to document on the data in any desired product that do not allow adequate end user-oriented calculation flexibility.
Share
It defines which the system tools all the security requirements for understanding and, if multiple write connection is needed, concurrent update location at an appropriated level, not all functions need customer to write data back, but for the increasing number which does, the system should be able to manage multiple updates in a timely, secure manner.
Multidimensional
This is the basic requirement. OLAP system must provide a multidimensional conceptual view of the data, including full support for hierarchies, as this is certainly the most logical method to analyze business and organizations.
Information
The system should be able to hold all the data needed by the applications. Data sparsity should be handled in an efficient manner.
The main characteristics of OLAP are as follows:
- Multidimensional conceptual view: OLAP systems let business users have a dimensional and logical view of the data in the data warehouse. It helps in carrying slice and dice operations.
- Multi-User Support: Since the OLAP techniques are shared, the OLAP operation should provide normal database operations, containing retrieval, update, adequacy control, integrity, and security.
- Accessibility: OLAP acts as a mediator between data warehouses and front-end. The OLAP operations should be sitting between data sources (e.g., data warehouses) and an OLAP front-end.
- Storing OLAP results: OLAP results are kept separate from data sources.
- Uniform documenting performance: Increasing the number of dimensions or database size should not significantly degrade the reporting performance of the OLAP system.
- OLAP provides for distinguishing between zero values and missing values so that aggregates are computed correctly.
- OLAP system should ignore all missing values and compute correct aggregate values.
- OLAP facilitate interactive query and complex analysis for the users.
- OLAP allows users to drill down for greater details or roll up for aggregations of metrics along a single business dimension or across multiple dimension.
- OLAP provides the ability to perform intricate calculations and comparisons.
- OLAP presents results in a number of meaningful ways, including charts and graphs.
Benefits of OLAP
OLAP holds several benefits for businesses: -
- OLAP helps managers in decision-making through the multidimensional record views that it is efficient in providing, thus increasing their productivity.
- OLAP functions are self-sufficient owing to the inherent flexibility support to the organized databases.
- It facilitates simulation of business models and problems, through extensive management of analysis-capabilities.
- In conjunction with data warehouse, OLAP can be used to support a reduction in the application backlog, faster data retrieval, and reduction in query drag.
Motivations for using OLAP
1) Understanding and improving sales: For enterprises that have much products and benefit a number of channels for selling the product, OLAP can help in finding the most suitable products and the most famous channels. In some methods, it may be feasible to find the most profitable users. For example, considering the telecommunication industry and considering only one product, communication minutes, there is a high amount of record if a company want to analyze the sales of products for every hour of the day (24 hours), difference between weekdays and weekends (2 values) and split regions to which calls are made into 50 region.
2) Understanding and decreasing costs of doing business: Improving sales is one method of improving a business, the other method is to analyze cost and to control them as much as suitable without affecting sales. OLAP can assist in analyzing the costs related to sales. In some methods, it may also be feasible to identify expenditures which produce a high return on investments (ROI). For example, recruiting a top salesperson may contain high costs, but the revenue generated by the salesperson may justify the investment.
OLAP Operations in the Multidimensional Data Model
In the multidimensional model, the records are organized into various dimensions, and each dimension includes multiple levels of abstraction described by concept hierarchies. This organization support users with the flexibility to view data from various perspectives. A number of OLAP data cube operation exist to demonstrate these different views, allowing interactive queries and search of the record at hand. Hence, OLAP supports a user-friendly environment for interactive data analysis.
Consider the OLAP operations which are to be performed on multidimensional data. The figure shows data cubes for sales of a shop. The cube contains the dimensions, location, and time and item, where the location is aggregated with regard to city values, time is aggregated with respect to quarters, and an item is aggregated with respect to item types.
Roll-Up
The roll-up operation (also known as drill-up or aggregation operation) performs aggregation on a data cube, by climbing down concept hierarchies, i.e., dimension reduction. Roll-up is like zooming-out on the data cubes. Figure shows the result of roll-up operations performed on the dimension location. The hierarchy for the location is defined as the Order Street, city, province, or state, country. The roll-up operation aggregates the data by ascending the location hierarchy from the level of the city to the level of the country.
When a roll-up is performed by dimensions reduction, one or more dimensions are removed from the cube. For example, consider a sales data cube having two dimensions, location and time. Roll-up may be performed by removing, the time dimensions, appearing in an aggregation of the total sales by location, relatively than by location and by time.
Drill-Down
The drill-down operation (also called roll-down) is the reverse operation of roll-up. Drill-down is like zooming-in on the data cube. It navigates from less detailed record to more detailed data. Drill-down can be performed by either stepping down a concept hierarchy for a dimension or adding additional dimensions.
Figure shows a drill-down operation performed on the dimension time by stepping down a concept hierarchy which is defined as day, month, quarter, and year. Drill-down appears by descending the time hierarchy from the level of the quarter to a more detailed level of the month.
Because a drill-down adds more details to the given data, it can also be performed by adding a new dimension to a cube. For example, a drill-down on the central cubes of the figure can occur by introducing an additional dimension, such as a customer group.
Slice
A slice is a subset of the cubes corresponding to a single value for one or more members of the dimension. For example, a slice operation is executed when the customer wants a selection on one dimension of a three-dimensional cube resulting in a two-dimensional site. So, the Slice operations perform a selection on one dimension of the given cube, thus resulting in a subcube.
Here Slice is functioning for the dimensions "time" using the criterion time = "Q1".
It will form a new sub-cubes by selecting one or more dimensions.
Dice
The dice operation describes a subcube by operating a selection on two or more dimension.
The dice operation on the cubes based on the following selection criteria involves three dimensions.
- (location = "Toronto" or "Vancouver")
- (time = "Q1" or "Q2")
- (item =" Mobile" or "Modem")
Pivot
The pivot operation is also called a rotation. Pivot is a visualization operations which rotates the data axes in view to provide an alternative presentation of the data. It may contain swapping the rows and columns or moving one of the row-dimensions into the column dimensions.
Other OLAP Operations
executes queries containing more than one fact table. The drill-through operations make use of relational SQL facilitates to drill through the bottom level of a data cubes down to its back-end relational tables.
Other OLAP operations may contain ranking the top-N or bottom-N elements in lists, as well as calculate moving average, growth rates, and interests, internal rates of returns, depreciation, currency conversions, and statistical tasks.
OLAP offers analytical modeling capabilities, containing a calculation engine for determining ratios, variance, etc. and for computing measures across various dimensions. It can generate summarization, aggregation, and hierarchies at each granularity level and at every dimensions intersection. OLAP also provide functional models for forecasting, trend analysis, and statistical analysis. In this context, the OLAP engine is a powerful data analysis tool.
UNIT 1B
Introduction
Data Mining Concept:
Why Data Mining?
“We are living in the information age” is a popular saying; however, we are actually living in the data age. Terabytes or petabytes (A petabyte is a unit of information or computer storage equal to 1 quadrillion bytes, or a thousand terabytes, or 1 million gigabytes) of data pour into our computer networks, the World Wide Web (WWW), and various data storage devices every day from business, society, science and engineering, medicine, and almost every other aspect of daily life. This explosive growth of available data volume is a result of the computerization of our society and the fast development of powerful data collection and storage tools. Businesses worldwide generate gigantic data sets, including sales transactions, stock trading records, product descriptions, sales promotions, company profiles and performance, and customer feedback. For example, large stores, such as Wal-Mart, handle hundreds of millions of transactions per week at thousands of branches around the world. Scientific and engineering practices generate high orders of petabytes of data in a continuous manner, from remote sensing, process measuring, scientific experiments, system performance, engineering observations, and environment surveillance. Global backbone telecommunication networks carry tens of petabytes of data traffic every day. The medical and health industry generates tremendous amounts of data from medical records, patient monitoring, and medical imaging. Billions of Web searches supported by search engines process tens of petabytes of data daily. Communities and social media have become increasingly important data sources, producing digital pictures and videos, blogs, Web communities, and various kinds of social networks. The list of sources that generate huge amounts of data is endless. This explosively growing, widely available, and gigantic body of data makes our time truly the data age. Powerful and versatile tools are badly needed to automatically uncover valuable information from the tremendous amounts of data and to transform such data into organized knowledge. This necessity has led to the birth of data mining.
What Is Data Mining?
It is no surprise that data mining, as a truly interdisciplinary subject, can be defined in many different ways. Even the term data mining does not really present all the major components in the picture. To refer to the mining of gold from rocks or sand, we say gold mining instead of rock or sand mining. Analogously, data mining should have been more appropriately named “knowledge mining from data,” which is unfortunately somewhat long. However, the shorter term, knowledge mining may not reflect the emphasis on mining from large amounts of data. Nevertheless, mining is a vivid term characterizing the process that finds a small set of precious nuggets from a great deal of raw material (Figure 1). Thus, such a misnomer carrying both “data” and “mining” became a popular choice. In addition, many other terms have a similar meaning to data mining—for example, knowledge mining from data, knowledge extraction, data/pattern analysis, data archaeology, and data dredging. Many people treat data mining as a synonym for another popularly used term, knowledge discovery from data, or KDD, while others view data mining as merely an essential step in the process of knowledge discovery.
The knowledge discovery process is shown in Figure as an iterative sequence of the following steps:
- Data cleaning (to remove noise and inconsistent data)
- Data integration (where multiple data sources may be combined)
- Data selection (where data relevant to the analysis task are retrieved from the database)
- Data transformation (where data are transformed and consolidated into forms appropriate for mining by performing summary or aggregation operations)
- Data mining (an essential process where intelligent methods are applied to extract data patterns)
- Pattern evaluation (to identify the truly interesting patterns representing knowledge based on interestingness measures)
- Knowledge presentation (where visualization and knowledge representation techniques are used to present mined knowledge to users)
Steps 1 through 4 are different forms of data preprocessing, where data are prepared for mining. The data mining step may interact with the user or a knowledge base. The interesting patterns are presented to the user and may be stored as new knowledge in the knowledge base. The preceding view shows data mining as one step in the knowledge discovery process, although an essential one because it uncovers hidden patterns for evaluation. However, in industry, in media, and in the research area, the term data mining is often used to refer to the entire knowledge discovery process (perhaps because the term is shorter than knowledge discovery from data). Therefore, we adopt a broad view of data mining functionality: Data mining is the process of discovering interesting patterns and knowledge from large amounts of data. The data sources can include databases, data warehouses, the Web, other information repositories, or data that are streamed into the system dynamically.
Data Mining Techniques
As a highly application-driven domain, data mining has incorporated many techniques from other domains such as statistics, machine learning, pattern recognition, database and data warehouse systems, information retrieval, visualization, algorithms, high performance computing, and many application domains. The interdisciplinary nature of data mining research and development contributes significantly to the success of data mining and its extensive applications.
Statistics
Statistics studies the collection, analysis, interpretation or explanation, and presentation of data. A statistical model is a set of mathematical functions that describe the behavior of the objects in a target class in terms of random variables and their associated probability distributions. For example, in data mining tasks like data characterization and classification, statistical models of target classes can be built. Alternatively, data mining tasks can be built on top of statistical models. For example, we can use statistics to model noise and missing data values. Then, when mining patterns in a large data set, the data mining process can use the model to help identify and handle noisy or missing values in the data. Statistical methods can be used to summarize or describe a collection of data. Inferential statistics (or predictive statistics) models data in a way that accounts for randomness and uncertainty in the observations and is used to draw inferences about the process or population under investigation. Statistical methods can also be used to verify data mining results. For example, after a classification or prediction model is mined, the model should be verified by statistical hypothesis testing.
Machine Learning
Machine learning is a fast-growing discipline. Here, we illustrate classic problems in machine learning that are highly related to data mining.
Supervised learning is basically a synonym for classification. The supervision in the learning comes from the labeled examples in the training data set. For example, in the postal code recognition problem, a set of handwritten postal code images and their corresponding machine-readable translations are used as the training examples, which supervise the learning of the classification model.
Unsupervised learning is essentially a synonym for clustering. The learning process is unsupervised since the input examples are not class labeled. Typically, we may use clustering to discover classes within the data.
Semi-supervised learning is a class of machine learning techniques that make use of both labeled and unlabeled examples when learning a model. In one approach, labeled examples are used to learn class models and unlabeled examples are used to refine the boundaries between classes. For a two-class problem, we can think of the set of examples belonging to one class as the positive examples and those belonging to the other class as the negative examples. In Figure, if we do not consider the unlabeled examples, the dashed line is the decision boundary that best partitions the positive examples from the negative examples. Using the unlabeled examples, we can refine the decision boundary to the solid line. Moreover, we can detect that the two positive examples at the top right corner, though labeled, are likely noise or outliers.
Active learning is a machine learning approach that lets users play an active role in the learning process. The goal is to optimize the model quality by actively acquiring knowledge from human users, given a constraint on how many examples they can be asked to label.
Database Systems and DataWarehouses
Many data mining tasks need to handle large data sets or even real-time, fast streaming data. Therefore, data mining can make good use of scalable database technologies to achieve high efficiency and scalability on large data sets.
Information Retrieval
Information retrieval (IR) is the science of searching for documents or information in documents. Documents can be text or multimedia, and may reside on the Web. The differences between traditional information retrieval and database systems are twofold: Information retrieval assumes that (1) the data under search are unstructured; and (2) the queries are formed mainly by keywords, which do not have complex structures (unlike SQL queries in database systems).
Major Issues in Data Mining
1] Data Quality
The quality of data used in data mining is one of the most significant challenges. The accuracy, completeness, and consistency of the data affect the accuracy of the results obtained. The data may contain errors, omissions, duplications, or inconsistencies, which may lead to inaccurate results. Moreover, the data may be incomplete, meaning that some attributes or values are missing, making it challenging to obtain a complete understanding of the data.
Data quality issues can arise due to a variety of reasons, including data entry errors, data storage issues, data integration problems, and data transmission errors. To address these challenges, data mining practitioners must apply data cleaning and data preprocessing techniques to improve the quality of the data. Data cleaning involves detecting and correcting errors, while data preprocessing involves transforming the data to make it suitable for data mining.
2] Data Complexity
Data complexity refers to the vast amounts of data generated by various sources, such as sensors, social media, and the internet of things (IoT). The complexity of the data may make it challenging to process, analyze, and understand. In addition, the data may be in different formats, making it challenging to integrate into a single dataset.
To address this challenge, data mining practitioners use advanced techniques such as clustering, classification, and association rule mining. These techniques help to identify patterns and relationships in the data, which can then be used to gain insights and make predictions.
3] Data Privacy and Security
Data privacy and security is another significant challenge in data mining. As more data is collected, stored, and analyzed, the risk of data breaches and cyber-attacks increases. The data may contain personal, sensitive, or confidential information that must be protected. Moreover, data privacy regulations such as GDPR, CCPA, and HIPAA impose strict rules on how data can be collected, used, and shared.
To address this challenge, data mining practitioners must apply data anonymization and data encryption techniques to protect the privacy and security of the data. Data anonymization involves removing personally identifiable information (PII) from the data, while data encryption involves using algorithms to encode the data to make it unreadable to unauthorized users.
4] Scalability
Data mining algorithms must be scalable to handle large datasets efficiently. As the size of the dataset increases, the time and computational resources required to perform data mining operations also increase. Moreover, the algorithms must be able to handle streaming data, which is generated continuously and must be processed in real-time.
To address this challenge, data mining practitioners use distributed computing frameworks such as Hadoop and Spark. These frameworks distribute the data and processing across multiple nodes, making it possible to process large datasets quickly and efficiently.
5] Interpretability
Data mining algorithms can produce complex models that are difficult to interpret. This is because the algorithms use a combination of statistical and mathematical techniques to identify patterns and relationships in the data. Moreover, the models may not be intuitive, making it challenging to understand how the model arrived at a particular conclusion.
To address this challenge, data mining practitioners use visualization techniques to represent the data and the models visually. Visualization makes it easier to understand the patterns and relationships in the data and to identify the most important variables.
6] Ethics
Data mining raises ethical concerns related to the collection, use, and dissemination of data. The data may be used to discriminate against certain groups, violate privacy rights, or perpetuate existing biases. Moreover, data mining algorithms may not be transparent, making it challenging to detect biases or discrimination.
Which Kinds of Applications Are Targeted?
1. Business Intelligence
Business intelligence (BI) technologies provide historical, current, and predictive views of business operations. Without data mining, many businesses may not be able to perform effective market analysis, compare customer feedback on similar products, discover the strengths and weaknesses of their competitors, retain highly valuable customers, and make smart business decisions. Clearly, data mining is the core of business intelligence.
2. Web Search Engines
A Web search engine is a specialized computer server that searches for information on the Web. The search results of a user query are often returned as a list (sometimes called hits). The hits may consist of web pages, images, and other types of files. Web search engines are essentially very large data mining applications. Various data mining techniques are used in all aspects of search engines, ranging from crawling (e.g., deciding which pages should be crawled and the crawling frequencies), indexing (e.g., selecting pages to be indexed and deciding to which extent the index should be constructed), and searching (e.g., deciding how pages should be ranked, which advertisements should be added, and how the search results can be personalized or made “context aware”).
Data Objects and Attribute Types
Data sets are made up of data objects. A data object represents an entity—in a sales database, the objects may be customers, store items, and sales; in a medical database, the objects may be patients; in a university database, the objects may be students, professors, and courses. Data objects are typically described by attributes. If the data objects are stored in a database, they are data tuples. That is, the rows of a database correspond to the data objects, and the columns correspond to the attributes.
What Is an Attribute?
An attribute is a data field, representing a characteristic or feature of a data object. The nouns attribute, dimension, feature, and variable are often used interchangeably in the literature. The term dimension is commonly used in data warehousing. Machine learning literature tends to use the term feature, while statisticians prefer the term variable. Data mining and database professionals commonly use the term attribute. Attributes describing a customer object can include, for example, customer ID, name, and address. A set of attributes used to describe a given object is called an attribute vector (or feature vector). The distribution of data involving one attribute (or variable) is called univariate. A bivariate distribution involves two attributes, and so on.
The type of an attribute is determined by the set of possible values—nominal, binary, ordinal, or numeric—the attribute can have.
1. Nominal Attributes
The values of a nominal attribute are symbols or names of things. Each value represents some kind of category, code, or state, and so nominal attributes are also referred to as categorical. The values do not have any meaningful order.
Example: Hair colour and marital status are two attributes describing person objects. The possible values for hair colour are black, brown, blond, red, auburn, grey, and white. The attribute marital status can take on the values single, married, divorced, and widowed. Both hair colour and marital status are nominal attributes. Another example of a nominal attribute is occupation, with the values teacher, dentist, programmer, farmer, and so on.
2. Binary Attributes
A binary attribute is a nominal attribute with only two categories or states: 0 or 1, where 0 typically means that the attribute is absent, and 1 means that it is present. Binary attributes are referred to as Boolean if the two states correspond to true and false.
Example: Given the attribute smoker describing a patient object, 1 indicates that the patient smokes, while 0 indicates that the patient does not.
A binary attribute is symmetric if both of its states are equally valuable and carry the same weight; that is, there is no preference on which outcome should be coded as 0 or 1. One such example could be the attribute gender having the states male and female.
A binary attribute is asymmetric if the outcomes of the states are not equally important, such as the positive and negative outcomes of a medical test for HIV. By convention, we code the most important outcome, which is usually the rarest one, by 1 (e.g., HIV positive) and the other by 0 (e.g., HIV negative).
3. Ordinal Attributes
An ordinal attribute is an attribute with possible values that have a meaningful order or ranking among them, but the magnitude between successive values is not known.
Example: Customer satisfaction had the following ordinal categories: 0: very dissatisfied, 1: somewhat dissatisfied, 2: neutral, 3: satisfied, and 4: very satisfied. Drink size corresponds to the size of drinks available at a fast-food restaurant. This nominal attribute has three possible values: small, medium, and large.
Note that nominal, binary, and ordinal attributes are qualitative. That is, they describe a feature of an object without giving an actual size or quantity. The values of such qualitative attributes are typically words representing categories. The central tendency of an ordinal attribute can be represented by its mode and its median (the middle value in an ordered sequence), but the mean cannot be defined.
Numeric Attributes
A numeric attribute is quantitative; that is, it is a measurable quantity, represented in integer or real values. Numeric attributes can be interval-scaled or ratio-scaled.
1. Interval-Scaled Attributes
Interval-scaled attributes are measured on a scale of equal-size units. The values of interval-scaled attributes have order and can be positive, 0, or negative. A temperature attribute is interval-scaled.
Example: Temperatures in Celsius and Fahrenheit do not have a true zero-point, that is, neither 0 ◦C nor 0◦F indicates “no temperature.” Although we can compute the difference between temperature values, we cannot talk of one temperature value as being a multiple of another. Without a true zero, we cannot say, for instance, that 10◦C is twice as warm as 5◦C. That is, we cannot speak of the values in terms of ratios. Similarly, there is no true zero-point for calendar dates. This brings us to ratio-scaled attributes, for which a true zero-point exits.
Because interval-scaled attributes are numeric, we can compute their mean value, in addition to the median and mode measures of central tendency.
2. Ratio-Scaled Attributes
A ratio-scaled attribute is a numeric attribute with an inherent zero-point. That is, if a measurement is ratio-scaled, we can speak of a value as being a multiple (or ratio) of another value.
Unlike temperatures in Celsius and Fahrenheit, the Kelvin (K) temperature scale has what is considered a true zero-point (0◦K = −273.15◦C): It is the point at which the particles that comprise matter have zero kinetic energy. Other examples include count attributes such as years of experience (e.g., the objects are employees) and number of words (e.g., the objects are documents). Additional examples include attributes to measure weight, height, latitude and longitude coordinates (e.g., when clustering houses), and monetary quantities (e.g., you are 100 times richer with $100 than with $1).
Discrete versus Continuous Attributes
Classification algorithms developed from the field of machine learning often talk of attributes as being either discrete or continuous.
A discrete attribute has a finite or countably infinite set of values, which may or may not be represented as integers. The attributes hair colour, smoker, medical test, and drink size each have a finite number of values, and so are discrete. If an attribute is not discrete, it is continuous. Continuous attributes are typically represented as floating-point variables.
Basic Statistical Descriptions of Data
For data preprocessing to be successful, it is essential to have an overall picture of your data. Basic statistical descriptions can be used to identify properties of the data and highlight which data values should be treated as noise or outliers. We start with measures of central tendency, which measure the location of the middle or center of a data distribution. Intuitively speaking, given an attribute, where do most of its values fall? In particular, we discuss the mean, median, mode, and midrange. In addition to assessing the central tendency of our data set, we also would like to have an idea of the dispersion of the data. That is, how are the data spread out. The most common data dispersion measures are the range, quartiles, and interquartile range; the five-number summary and boxplots; and the variance and standard deviation of the data These measures are useful for identifying outliers.
Measuring the Central Tendency: Mean, Median, and Mode
The most common and effective numeric measure of the “center” of a set of data is the (arithmetic) mean. Let x1,x2,...,xN be a set of N values or observations, such as for some numeric attribute X, like salary. The mean of this set of values is
Sometimes, each value xi in a set may be associated with a weight wi for i = 1,...,N. The weights reflect the significance, importance, or occurrence frequency attached to their respective values. In this case, we can compute
This is called the weighted arithmetic mean or the weighted average.
Although the mean is the single most useful quantity for describing a data set, it is not always the best way of measuring the center of the data. A major problem with the mean is its sensitivity to extreme (e.g., outlier) values. Even a small number of extreme values can corrupt the mean. For skewed (asymmetric) data, a better measure of the center of data is the median, which is the middle value in a set of ordered data values. It is the value that separates the higher half of a data set from the lower half.
In probability and statistics, the median generally applies to numeric data; however, we may extend the concept to ordinal data. Suppose that a given data set of N values for an attribute X is sorted in increasing order. If N is odd, then the median is the middle value of the ordered set. If N is even, then the median is not unique; it is the two middlemost values and any value in between. If X is a numeric attribute in this case, by convention, the median is taken as the average of the two middlemost values.
where L1 is the lower boundary of the median interval, N is the number of values in the entire data set, (∑freq) l is the sum of the frequencies of all of the intervals that are lower than the median interval, freq(median) is the frequency of the median interval, and width is the width of the median interval.
The mode is another measure of central tendency. The mode for a set of data is the value that occurs most frequently in the set. Therefore, it can be determined for qualitative and quantitative attributes. It is possible for the greatest frequency to correspond to several different values, which results in more than one mode. Data sets with one, two, or three modes are respectively called unimodal, bimodal, and trimodal. In general, a data set with two or more modes is multimodal. At the other extreme, if each data value occurs only once, then there is no mode.
In a unimodal frequency curve with perfect symmetric data distribution, the mean, median, and mode are all at the same center value, as shown in Figure 2.1(a). Data in most real applications are not symmetric. They may instead be either positively skewed, where the mode occurs at a value that is smaller than the median (Figure 2.1b), or negatively skewed, where the mode occurs at a value greater than the median (Figure 2.1c).
Measuring the Dispersion of Data: Range, Quartiles, Variance, Standard Deviation, and Interquartile Range
We now look at measures to assess the dispersion or spread of numeric data. The measures include range, quantiles, quartiles, percentiles, and the interquartile range. Variance and standard deviation also indicate the spread of a data distribution.
Range, Quartiles, and Interquartile Range :
Let x1, x2, …, xN be a set of observations for some numeric attribute, X. The range of the set is the difference between the largest (max ()) and smallest (min ()) values.
Suppose that the data for attribute X are sorted in increasing numeric order. Imagine that we can pick certain data points so as to split the data distribution into equal-size consecutive sets, as in Figure. These data points are called quantiles. Quantiles are points taken at regular intervals of a data distribution, dividing it into essentially equalsize consecutive sets.
The 2-quantile is the data point dividing the lower and upper halves of the data distribution. It corresponds to the median. The 4-quantiles are the three data points that split the data distribution into four equal parts; each part represents one-fourth of the data distribution. They are more commonly referred to as quartiles. The 100-quantiles are more commonly referred to as percentiles; they divide the data distribution into 100 equal-sized consecutive sets. The median, quartiles, and percentiles are the most widely used forms of quantiles.
The quartiles give an indication of a distribution’s center, spread, and shape. The first quartile, denoted by Q1, is the 25th percentile. It cuts off the lowest 25% of the data. The third quartile, denoted by Q3, is the 75th percentile—it cuts off the lowest 75% (or highest 25%) of the data. The second quartile is the 50th percentile. As the median, it gives the center of the data distribution.The distance between the first and third quartiles is a simple measure of spread that gives the range covered by the middle half of the data. This distance is called the interquartile range (IQR) and is defined as
IQR = Q3 - Q1
Five-Number Summary, Boxplots, and Outliers:
No single numeric measure of spread (e.g., IQR) is very useful for describing skewed distributions. A common rule of thumb for identifying suspected outliers is to single out values falling at least 1.5 x IQR above the third quartile or below the first quartile.
Because Q1, the median, and Q3 together contain no information about the endpoints (e.g., tails) of the data, a fuller summary of the shape of a distribution can be obtained by providing the lowest and highest data values as well. This is known as the five-number summary. The five-number summary of a distribution consists of the median (Q2), the quartiles Q1 and Q3, and the smallest and largest individual observations, written in the order of Minimum, Q1, Median, Q3, Maximum.
Boxplots are a popular way of visualizing a distribution.
- Typically, the ends of the box are at the quartiles so that the box length is the interquartile range.
- The median is marked by a line within the box.
- Two lines (called whiskers) outside the box extend to the smallest (Minimum) and largest (Maximum) observations.
Variance and Standard Deviation:
Variance and standard deviation are measures of data dispersion. They indicate how spread out a data distribution is. A low standard deviation means that the data observations tend to be very close to the mean, while a high standard deviation indicates that the data are spread out over a large range of values.
The variance of N observations, x1, x2, . . ., xN, for a numeric attribute X is
Data Preprocessing
Data Quality: Why Preprocess the Data?
Data have quality if they satisfy the requirements of the intended use. There are many factors comprising data quality, including accuracy, completeness, consistency, timeliness, believability, and interpretability.
Inaccurate, incomplete, and inconsistent data are commonplace properties of large real-world databases and data warehouses. There are many possible reasons for inaccurate data. The data collection instruments used may be faulty. There may have been human or computer errors occurring at data entry. Users may purposely submit incorrect data values for mandatory fields when they do not wish to submit personal information. This is known as disguised missing data. Errors in data transmission can also occur. There may be technology limitations. Incorrect data may also result from inconsistencies in naming conventions or data codes, or inconsistent formats for input fields (e.g., date). Duplicate tuples also require data cleaning.
Incomplete data can occur for a number of reasons. Attributes of interest may not always be available, such as customer information for sales transaction data. Other data may not be included simply because they were not considered important at the time of entry. Relevant data may not be recorded due to a misunderstanding or because of equipment malfunctions. Data that were inconsistent with other recorded data may have been deleted. Furthermore, the recording of the data history or modifications may have been overlooked.
Timelines also affects data quality. Suppose in a company several sales representatives fail to submit their sales records on time at the end of the month. There are also a number of corrections and adjustments that flow in after the month’s end. So, the data stored in the database are incomplete. However, once all of the data are received, it is correct. The fact that the month-end data are not updated in a timely fashion has a negative impact on the data quality.
Believability reflects how much the data are trusted by users, while interpretability reflects how easy the data are understood. Suppose that a database, at one point, had several errors, all of which have since been corrected. The past errors, however, had caused many problems for sales department users, and so they no longer trust the data. The data also use many accounting codes, which the sales department does not know how to interpret. Even though the database is now accurate, complete, consistent, and timely, sales department users may regard it as of low quality due to poor believability and interpretability.
Data Cleaning
Real-world data tend to be incomplete, noisy, and inconsistent. Data cleaning attempt to fill in missing values, smooth out noise while identifying outliers, and correct inconsistencies in the data. In this section, you will study basic methods for data cleaning.
- Missing Values
How can you go about filling in the missing values for an attribute?
a) Ignore the tuple: This is usually done when the class label is missing (assuming the mining task involves classification). This method is not very effective, unless the tuple contains several attributes with missing values.
By ignoring the tuple, we do not make use of the remaining attributes’ values in the tuple. Such data could have been useful to the task at hand.
b) Fill in the missing value manually: In general, this approach is time consuming and may not be feasible given a large data set with many missing values.
c) Use a global constant to fill in the missing value: Replace all missing attribute values by the same constant such as a label like “Unknown”. If missing values are replaced by, say, “Unknown,” then the mining program may mistakenly think that they form an interesting concept, since they all have a value in common—that of “Unknown.” Hence, although this method is simple, it is not foolproof.
d) Use a measure of central tendency for the attribute (e.g., the mean or median) to fill in the missing value: Measures of central tendency, which indicate the “middle” value of a data distribution. For normal (symmetric) data distributions, the mean can be used, while skewed data distribution should employ the median.
e) Use the most probable value to fill in the missing value: This may be determined with regression, inference-based tools using a Bayesian formalism, or decision tree induction.
- Noisy Data
Noise is a random error or variance in a measured variable. Outliers may represent noise.
a) Binning: Binning methods smooth a sorted data value by consulting the values around it. The sorted values are distributed into a number of “buckets,” or bins. Because binning methods consult the neighbourhood of values, they perform local smoothing.
In this example, the data for price are first sorted and then partitioned into equal-frequency bins of size 3 (i.e., each bin contains three values).
Some binning techniques:
- In smoothing by bin means, each value in a bin is replaced by the mean value of the bin.
- Similarly, smoothing by bin medians can be employed, in which each bin value is replaced by the bin median.
- In smoothing by bin boundaries, the minimum and maximum values in a given bin are identified as the bin boundaries. Each bin value is then replaced by the closest boundary value.
b) Regression: Data smoothing can also be done by regression, a technique that conforms data values to a function. Linear regression involves finding the “best” line to fit two attributes (or variables) so that one attribute can be used to predict the other. Multiple linear regression is an extension of linear regression, where more than two attributes are involved and the data are fit to a multidimensional surface.
c) Outlier analysis: Outliers may be detected by clustering, for example, where similar values are organized into groups, or “clusters.” Intuitively, values that fall outside of the set of clusters may be considered outliers.
Data Cleaning as a Process
The first step in data cleaning as a process is discrepancy detection.
“So, how can we proceed with discrepancy detection?” As a starting point, use any knowledge you may already have regarding properties of the data. Such knowledge or “data about data” is referred to as metadata. This is where we can make use of metadata. what are the data type and domain of each attribute? What are the acceptable values for each attribute? The basic statistical data descriptions are useful here to grasp data trends and identify anomalies. For example, find the mean, median, and mode values. Are the data symmetric or skewed? What is the range of values? Do all values fall within the expected range? What is the standard deviation of each attribute? Values that are more than two standard deviations away from the mean for a given attribute may be flagged as potential outliers. Are there any known dependencies between attributes?
The data should also be examined regarding unique rules, consecutive rules, and null rules. A unique rule says that each value of the given attribute must be different from all other values for that attribute. A consecutive rule says that there can be no missing values between the lowest and highest values for the attribute, and that all values must also be unique. A null rule specifies the use of blanks, question marks, special characters, or other strings that may indicate the null condition and how such values should be handled. Data scrubbing tools use simple domain knowledge (e.g., knowledge of postal addresses and spell-checking) to detect errors and make corrections in the data. Data auditing tools find discrepancies by analysing the data to discover rules and relationships, and detecting data that violate such conditions. Some data inconsistencies may be corrected manually using external references.
Data Integration
Data mining often requires data integration—the merging of data from multiple data stores. Careful integration can help reduce and avoid redundancies and inconsistencies in the resulting data set. This can help improve the accuracy and speed of the subsequent data mining process.
- Entity Identification Problem
It is likely that your data analysis task will involve data integration, which combines data from multiple sources into a coherent data store, as in data warehousing. These sources may include multiple databases, data cubes, or flat files.
There are a number of issues to consider during data integration. Schema integration and object matching can be tricky. How can equivalent real-world entities from multiple data sources be matched up? This is referred to as the entity identification problem. For example, how can the data analyst or the computer be sure that customer id in one database and cust number in another refer to the same attribute?
- Redundancy and Correlation Analysis
Redundancy is another important issue in data integration. Some redundancies can be detected by correlation analysis. Given two attributes, such analysis can measure how strongly one attribute implies the other, based on the available data. For nominal data, we use the chi-square test. For numeric attributes, we can use the correlation coefficient and covariance, both of which access how one attribute’s values vary from those of another.
a) Correlation Test for Nominal Data
b) Correlation Coefficient for Numeric Data
For numeric attributes, we can evaluate the correlation between two attributes, A and B, by computing the correlation coefficient (also known as Pearson’s product moment coefficient, named after its inventor, Karl Pearson). This is
If the resulting value is equal to 0, then A and B are independent and there is no correlation between them. If the resulting value is less than 0, then A and B are negatively correlated, where the values of one attribute increase as the values of the other attribute decrease. This means that each attribute discourages the other.
c) Covariance of Numeric Data
In probability theory and statistics, correlation and covariance are two similar measures for assessing how much two attributes change together. Consider two numeric attributes A and B, and a set of n observations {(a1,b1), . . . , (an,bn)}. The mean values of A and B, respectively, are also known as the expected values on A and B, that is,
- Tuple Duplication
In addition to detecting redundancies between attributes, duplication should also be detected at the tuple level (e.g., where there are two or more identical tuples for a given unique data entry case). The use of denormalized tables (often done to improve performance by avoiding joins) is another source of data redundancy. Inconsistencies often arise between various duplicates, due to inaccurate data entry or updating some but not all data occurrences.
- Data Value Conflict Detection and Resolution
Data integration also involves the detection and resolution of data value conflicts. For example, for the same real-world entity, attribute values from different sources may differ. This may be due to differences in representation, scaling, or encoding.
Data Reduction
Data reduction techniques can be applied to obtain a reduced representation of the data set that is much smaller in volume, yet closely maintains the integrity of the original data. That is, mining on the reduced data set should be more efficient yet produce the same (or almost the same) analytical results.
Data reduction strategies include dimensionality reduction, numerosity reduction, and data compression.
Dimensionality reduction is the process of reducing the number of random variables or attributes under consideration. Dimensionality reduction methods include wavelet transforms and principal components analysis, which transform or project the original data onto a smaller space. Attribute subset selection is a method of dimensionality reduction in which irrelevant, weakly relevant, or redundant attributes or dimensions are detected and removed.
Numerosity reduction techniques replace the original data volume by alternative, smaller forms of data representation. These techniques may be parametric or nonparametric.
For parametric methods, a model is used to estimate the data, so that typically, only the data parameters need to be stored, instead of the actual data. (Outliers may also be stored.) Regression and log-linear models are examples.
Nonparametric methods for storing reduced representations of the data include histograms, clustering, sampling, and data cube aggregation.
In data compression, transformations are applied so as to obtain a reduced or “compressed” representation of the original data. If the original data can be reconstructed from the compressed data without any information loss, the data reduction is called lossless. If, instead, we can reconstruct only an approximation of the original data, then the data reduction is called lossy.
Data Transformation
This section presents methods of data transformation. In this preprocessing step, the data are transformed or consolidated so that the resulting mining process may be more efficient, and the patterns found may be easier to understand.
Data Transformation Strategies Overview
In data transformation, the data are transformed or consolidated into forms appropriate for mining. Strategies for data transformation include the following:
- Smoothing, which works to remove noise from the data. Techniques include binning, regression, and clustering.
- Attribute construction (or feature construction), where new attributes are constructed and added from the given set of attributes to help the mining process.
- Aggregation, where summary or aggregation operations are applied to the data. For example, the daily sales data may be aggregated so as to compute monthly and annual total amounts.
- Normalization, where the attribute data are scaled so as to fall within a smaller range, such as −1.0 to 1.0, or 0.0 to 1.0.
- Discretization, where the raw values of a numeric attribute (e.g., age) are replaced by interval labels (e.g., 0–10, 11–20, etc.) or conceptual labels (e.g., youth, adult, senior).
- Concept hierarchy generation for nominal data, where attributes such as street can be generalized to higher-level concepts, like city or country.
Data Transformation by Normalization
There are many methods for data normalization. We study min-max normalization, z-score normalization, and normalization by decimal scaling. For our discussion, let A be a numeric attribute with n observed values, v1, v2,..., vn.
Min-max normalization performs a linear transformation on the original data. Suppose that minA and maxA are the minimum and maximum values of an attribute, A.
In z-score normalization (or zero-mean normalization), the values for an attribute, A, are normalized based on the mean (i.e., average) and standard deviation of A. A value,
Data Visualization
Data visualization aims to communicate data clearly and effectively through graphical representation. Data visualization has been used extensively in many applications—for example, at work for reporting, managing business operations, and tracking progress of tasks. More popularly, we can take advantage of visualization techniques to discover data relationships that are otherwise not easily observable by looking at the raw data.
-
Pixel-Oriented Visualization Techniques
A simple way to visualize the value of a dimension is to use a pixel where the color of the pixel reflects the dimension’s value.
-
Geometric Projection Visualization Techniques
A drawback of pixel-oriented visualization techniques is that they cannot help us much in understanding the distribution of data in a multidimensional space. Geometric projection techniques help users find interesting projections of multidimensional data sets. The central challenge the geometric projection techniques try to address is how to visualize a high-dimensional space on a 2-D display. A scatter plot displays 2-D data points using Cartesian coordinates. A third dimension can be added using different colors or shapes to represent different data points. A 3-D scatter plot uses three axes in a Cartesian coordinate system. If it also uses color, it can display up to 4-D data points.
-
Icon-Based Visualization Techniques
Icon-based visualization techniques use small icons to represent multidimensional data values. We look at two popular icon-based techniques: Chernoff faces and stick figures. Chernoff faces were introduced in 1973 by statistician Herman Chernoff. They display multidimensional data of up to 18 variables (or dimensions) as a cartoon human face . Chernoff faces help reveal trends in the data. Components of the face, such as the eyes, ears, mouth, and nose, represent values of the dimensions by their shape, size, placement, and orientation. For example, dimensions can be mapped to the following facial characteristics: eye size, eye spacing, nose length, nose width, mouth curvature, mouth width, mouth openness, pupil size, eyebrow slant, eye eccentricity, and head eccentricity. Viewing large tables of data can be tedious. By condensing the data, Chernoff faces make the data easier for users to digest. In this way, they facilitate visualization of regularities and irregularities present in the data, although their power in relating multiple relationships is limited. Another limitation is that specific data values are not shown.
The stick figure visualization technique maps multidimensional data to five-piece stick figures, where each figure has four limbs and a body. Two dimensions are mapped to the display (x and y) axes and the remaining dimensions are mapped to the angle and/or length of the limbs. Figure shows census data, where age and income are mapped to the display axes, and the remaining dimensions (gender, education, and so on) are mapped to stick figures. If the data items are relatively dense with respect to the two display dimensions, the resulting visualization shows texture patterns, reflecting data trends.
-
Hierarchical Visualization Techniques
Hierarchical visualization techniques partition all dimensions into subsets (i.e., subspaces). The subspaces are visualized in a hierarchical manner. “Worlds-within-Worlds,” also known as n-Vision, is a representative hierarchical visualization method. As another example of hierarchical visualization methods, tree-maps display hierarchical data as a set of nested rectangles.
Measuring Data Similarity and Dissimilarity
In data mining applications, such as clustering, outlier analysis, and nearest-neighbour classification, we need ways to assess how alike or unalike objects are in comparison to one another. For example, a store may want to search for clusters of customer objects, resulting in groups of customers with similar characteristics (e.g., similar income, area of residence, and age). Such information can then be used for marketing. A cluster is a collection of data objects such that the objects within a cluster are similar to one another and dissimilar to the objects in other clusters. Outlier analysis also employs clustering-based techniques to identify potential outliers as objects that are highly dissimilar to others.
Similarity and dissimilarity measures, are also referred to as measures of proximity.
Data Matrix versus Dissimilarity Matrix
Data matrix (or object-by-attribute structure): This structure stores the n data objects in the form of a relational table, or n-by-p matrix (n objects X p attributes):
Dissimilarity matrix (or object-by-object structure): This structure stores a collection of proximities that are available for all pairs of n objects. It is often represented by an n-by-n table:
where d(i, j) is the measured dissimilarity or “difference” between objects i and j. In general, d(i, j) is a non-negative number that is close to 0 when objects i and j are highly similar or “near” each other, and becomes larger the more they differ. Note that d(i, i) = 0; that is, the difference between an object and itself is 0.
Measures of similarity can often be expressed as a function of measures of dissimilarity.
For example, for nominal data,
Sim(i, j) = 1-d(i, j)
Proximity Measures for Nominal Attributes
“How is dissimilarity computed between objects described by nominal attributes? “The dissimilarity between two objects i and j can be computed based on the ratio of mismatches:
D(i, j) =(p-m)/p
where m is the number of matches (i.e., the number of attributes for which i and j are in the same state), and p is the total number of attributes describing the objects. Weights can be assigned to increase the effect of m or to assign greater weight to the matches in attributes having a larger number of states.
Proximity Measures for Binary Attributes
“So, how can we compute the dissimilarity between two binary attributes?” One approach involves computing a dissimilarity matrix from the given binary data. If all binary attributes are thought of as having the same weight, we have the 2X2 contingency table. where q is the number of attributes that equal 1 for both objects i and j, r is the number of attributes that equal 1 for object i but equal 0 for object j, s is the number of attributes that equal 0 for object i but equal 1 for object j, and t is the number of attributes that equal 0 for both objects i and j. The total number of attributes is p, where p = q+r +s +t .
UNIT 2A
Data Mining, frequent pattern Analysis
Imagine that you are a sales manager at an Electronics store, and you are talking to a customer who recently bought a PC and a digital camera from the store. What should you recommend to her next? Information about which products are frequently purchased by your customers following their purchases of a PC and a digital camera in sequence would be very helpful in making your recommendation. Frequent patterns and association rules are the knowledge that you want to mine in such a scenario.
Frequent patterns are patterns (e.g., itemsets, subsequences, or substructures) that appear frequently in a data set. For example, a set of items, such as milk and bread, that appear frequently together in a transaction data set is a frequent itemset. A subsequence, such as buying first a PC, then a digital camera, and then a memory card, if it occurs frequently in a shopping history database, is a (frequent) sequential pattern.
Finding frequent patterns plays an essential role in mining associations, correlations, and many other interesting relationships among data. Moreover, it helps in data classification, clustering, and other data mining tasks. Thus, frequent pattern mining has become an important data mining task and a focused theme in data mining research.
Basic Concepts
Frequent pattern mining searches for recurring relationships in a given data set. This section introduces the basic concepts of frequent pattern mining for the discovery of interesting associations and correlations between itemsets in transactional and relational databases. We begin by presenting an example of market basket analysis, the earliest form of frequent pattern mining for association rules.
Market Basket Analysis: A Motivating Example
Frequent itemset mining leads to the discovery of associations and correlations among items in large transactional or relational data sets. With massive amounts of data continuously being collected and stored, many industries are becoming interested in mining such patterns from their databases. The discovery of interesting correlation relationships among huge amounts of business transaction records can help in many business decision-making processes such as catalog design, cross-marketing, and customer shopping behavior analysis.
A typical example of frequent itemset mining is market basket analysis. This process analyzes customer buying habits by finding associations between the different items that customers place in their “shopping baskets”. The discovery of these associations can help retailers develop marketing strategies by gaining insight into which items are frequently purchased together by customers. For instance, if customers are buying milk, how likely are they to also buy bread (and what kind of bread) on the same trip to the supermarket? This information can lead to increased sales by helping retailers do selective marketing and plan their shelf space.
If we think of the universe as the set of items available at the store, then each item has a Boolean variable representing the presence or absence of that item. Each basket can then be represented by a Boolean vector of values assigned to these variables. The Boolean vectors can be analyzed for buying patterns that reflect items that are frequently associated or purchased together. These patterns can be represented in the form of association rules. For example, the information that customers who purchase computers also tend to buy antivirus software at the same time is represented in the following association rule:
computer=>antivirus software [support= 2%, confidence = 60%]. ----(1)
Rule support and confidence are two measures of rule interestingness. A support of 2% for Rule (1) means that 2% of all the transactions under analysis show that computer and antivirus software are purchased together. A confidence of 60% means that 60% of the customers who purchased a computer also bought the software. Typically, association rules are considered interesting if they satisfy both a minimum support threshold and a minimum confidence threshold. These thresholds can be a set by users or domain experts. Additional analysis can be performed to discover interesting statistical correlations between associated items.
Frequent Itemset Mining Methods
We begin by presenting Apriori, the basic algorithm for finding frequent itemsets.
Apriori Algorithm: Finding Frequent Itemsets by Confined Candidate Generation**
Apriori is a seminal algorithm proposed by R. Agrawal and R. Srikant in 1994 for mining frequent itemsets for Boolean association rules. The name of the algorithm is based on the fact that the algorithm uses prior knowledge of frequent itemset properties, as we shall see later. Apriori employs an iterative approach known as a level-wise search, where k-itemsets are used to explore (k +1)-itemsets.
First, the set of frequent 1-itemsets is found by scanning the database to accumulate the count for each item, and collecting those items that satisfy minimum support. The resulting set is denoted by L1. Next, L1 is used to find L2, the set of frequent 2-itemsets, which is used to find L3, and so on, until no more frequent k-itemsets can be found. The finding of each Lk requires one full scan of the database.
To improve the efficiency of the level-wise generation of frequent itemsets, an important property called the Apriori property is used to reduce the search space.
Apriori property: All nonempty subsets of a frequent itemset must also be frequent.
The Apriori property is based on the following observation. By definition, if an itemset I does not satisfy the minimum support threshold, min sup, then I is not frequent, that is, P(I) < min sup. If an item A is added to the itemset I, then the resulting itemset (i.e., I U A) cannot occur more frequently than I. Therefore, I U A is not frequent either,that is, P(I U A) < min sup.
This property belongs to a special category of properties called antimonotonicity in the sense that if a set cannot pass a test, all of its supersets will fail the same test as well. It is called antimonotonicity because the property is monotonic in the context of failing a test.
Generating Association Rules from Frequent Itemsets
Once the frequent itemsets from transactions in a database D have been found, it is straightforward to generate strong association rules from them (where strong association rules satisfy both minimum support and minimum confidence). This can be done using Eq. for confidence, which we show again here for completeness:
confidence (A=>B) = P(B|A) =support (A U B)/ support (A) = support count(A U B)/ support count(A)
The conditional probability is expressed in terms of itemset support count, where support count(A U B) is the number of transactions containing the itemsets A U B, and support count(A) is the number of transactions containing the itemset A. Based on this equation, association rules can be generated as follows:
- For each frequent itemset l, generate all nonempty subsets of l.
- For every nonempty subset s of l, output the rule “s=>(l –s)” if support count(l)/support count(s)>=min conf, where min conf is the minimum confidence threshold.
Because the rules are generated from frequent itemsets, each one automatically satisfies the minimum support. Frequent itemsets can be stored ahead of time in hash tables along with their counts so that they can be accessed quickly.
FP Growth Algorithm in Data Mining
In Data Mining, finding frequent patterns in large databases is very important and has been studied on a large scale in the past few years. Unfortunately, this task is computationally expensive, especially when many patterns exist.
The FP-Growth Algorithm proposed by Han in. This is an efficient and scalable method for mining the complete set of frequent patterns by pattern fragment growth, using an extended prefix-tree structure for storing compressed and crucial information about frequent patterns named frequent-pattern tree (FP-tree). In his study, Han proved that his method outperforms other popular methods for mining frequent patterns, e.g. the Apriori Algorithm and the TreeProjection. In some later works, it was proved that FP-Growth performs better than other methods, including Eclat and Relim. The popularity and efficiency of the FP-Growth Algorithm contribute to many studies that propose variations to improve its performance.
What is FP Growth Algorithm?
The FP-Growth Algorithm is an alternative way to find frequent item sets without using candidate generations, thus improving performance. For so much, it uses a divide-and-conquer strategy. The core of this method is the usage of a special data structure named frequent-pattern tree (FP-tree), which retains the item set association information.
This algorithm works as follows:
- First, it compresses the input database creating an FP-tree instance to represent frequent items.
- After this first step, it divides the compressed database into a set of conditional databases, each associated with one frequent pattern.
- Finally, each such database is mined separately.
Using this strategy, the FP-Growth reduces the search costs by recursively looking for short patterns and then concatenating them into the long frequent patterns.
In large databases, holding the FP tree in the main memory is impossible. A strategy to cope with this problem is to partition the database into a set of smaller databases (called projected databases) and then construct an FP-tree from each of these smaller databases.
FP-Tree
The frequent-pattern tree (FP-tree) is a compact data structure that stores quantitative information about frequent patterns in a database. Each transaction is read and then mapped onto a path in the FP-tree. This is done until all transactions have been read. Different transactions with common subsets allow the tree to remain compact because their paths overlap.
A frequent Pattern Tree is made with the initial item sets of the database. The purpose of the FP tree is to mine the most frequent pattern. Each node of the FP tree represents an item of the item set.
The root node represents null, while the lower nodes represent the item sets. The associations of the nodes with the lower nodes, that is, the item sets with the other item sets, are maintained while forming the tree.
Han defines the FP-tree as the tree structure given below:
- One root is labelled as "null" with a set of item-prefix subtrees as children and a frequent-item-header table.
-
Each node in the item-prefix subtree consists of three fields:
- Item-name: registers which item is represented by the node;
- Count: the number of transactions represented by the portion of the path reaching the node;
- Node-link: links to the next node in the FP-tree carrying the same item name or null if there is none.
-
Each entry in the frequent-item-header table consists of two fields:
- Item-name: as the same to the node;
- Head of node-link: a pointer to the first node in the FP-tree carrying the item name.
Additionally, the frequent-item-header table can have the count support for an item. The below diagram is an example of a best-case scenario that occurs when all transactions have the same itemset; the size of the FP-tree will be only a single branch of nodes.
The worst-case scenario occurs when every transaction has a unique item set. So the space needed to store the tree is greater than the space used to store the original data set because the FP-tree requires additional space to store pointers between nodes and the counters for each item. The diagram below shows how a worst-case scenario FP-tree might appear. As you can see, the tree's complexity grows with each transaction's uniqueness.
Algorithm by Han
The original algorithm to construct the FP-Tree defined by Han is given below:
Algorithm 1: FP-tree construction
Input: A transaction database DB and a minimum support threshold?
Output: FP-tree, the frequent-pattern tree of DB.
Method: The FP-tree is constructed as follows.
- The first step is to scan the database to find the occurrences of the itemsets in the database. This step is the same as the first step of Apriori. The count of 1-itemsets in the database is called support count or frequency of 1-itemset.
- The second step is to construct the FP tree. For this, create the root of the tree. The root is represented by null.
- The next step is to scan the database again and examine the transactions. Examine the first transaction and find out the itemset in it. The itemset with the max count is taken at the top, and then the next itemset with the lower count. It means that the branch of the tree is constructed with transaction itemsets in descending order of count.
-
The next transaction in the database is examined. The itemsets are ordered in descending order of count. If any itemset of this transaction is already present in another branch, then this transaction branch would share a common prefix to the root.
This means that the common itemset is linked to the new node of another itemset in this transaction.
- Also, the count of the itemset is incremented as it occurs in the transactions. The common node and new node count are increased by 1 as they are created and linked according to transactions.
- The next step is to mine the created FP Tree. For this, the lowest node is examined first, along with the links of the lowest nodes. The lowest node represents the frequency pattern length 1. From this, traverse the path in the FP Tree. This path or paths is called a conditional pattern base.
A conditional pattern base is a sub-database consisting of prefix paths in the FP tree occurring with the lowest node (suffix). - Construct a Conditional FP Tree, formed by a count of itemsets in the path. The itemsets meeting the threshold support are considered in the Conditional FP Tree.
- Frequent Patterns are generated from the Conditional FP Tree.
Using this algorithm, the FP-tree is constructed in two database scans. The first scan collects and sorts the set of frequent items, and the second constructs the FP-Tree.
ECLAT Algorithm
The ECLAT algorithm stands for Equivalence Class Clustering and bottom-up Lattice Traversal. It is one of the popular methods of Association Rule mining. It is a more efficient and scalable version of the Apriori algorithm. While the Apriori algorithm works in a horizontal sense imitating the Breadth-First Search of a graph, the ECLAT algorithm works in a vertical manner just like the Depth-First Search of a graph. This vertical approach of the ECLAT algorithm makes it a faster algorithm than the Apriori algorithm.
How the algorithm work? :
The basic idea is to use Transaction Id Sets(tidsets) intersections to compute the support value of a candidate and avoiding the generation of subsets which do not exist in the prefix tree. In the first call of the function, all single items are used along with their tidsets. Then the function is called recursively and in each recursive call, each item-tidset pair is verified and combined with other item-tidset pairs. This process is continued until no candidate item-tidset pairs can be combined.
Advantages over Apriori algorithm:-
- Memory Requirements: Since the ECLAT algorithm uses a Depth-First Search approach, it uses less memory than Apriori algorithm.
- Speed: The ECLAT algorithm is typically faster than the Apriori algorithm.
- Number of Computations: The ECLAT algorithm does not involve the repeated scanning of the data to compute the individual support values.
Pattern Evaluation Methods
Most association rule mining algorithms employ a support–confidence framework. Although minimum support and confidence thresholds help weed out or exclude the exploration of a good number of uninteresting rules, many of the rules generated are still not interesting to the users. Unfortunately, this is especially true when mining at low support thresholds or mining for long patterns. This has been a major bottleneck for successful application of association rule mining.
1. Strong Rules Are Not Necessarily Interesting
Whether or not a rule is interesting can be assessed either subjectively or objectively. Ultimately, only the user can judge if a given rule is interesting, and this judgment, being subjective, may differ from one user to another.
A misleading “strong” association rule:
Suppose we are interested in analyzing transactions at AllElectronics with respect to the purchase of computer games and videos. Let game refer to the transactions containing computer games, and video refer to those containing videos. Of the 10,000 transactions analyzed, the data show that 6000 of the customer transactions included computer games, while 7500 included videos, and 4000 included both computer games and videos. Suppose that a data mining program for discovering association rules is run on the data, using a minimum support of, say, 30% and a minimum confidence of 60%. The following association rule is discovered:
Rule (6.6) is a strong association rule and would therefore be reported, since its support value of 4000/10,000 = 40% and confidence value of 4000/6000
= 66% satisfy the minimum support and minimum confidence thresholds, respectively. However, Rule (6.6) is misleading because the probability of purchasing videos is 75%, which is even larger than 66%. In fact, computer games and videos are negatively associated because the purchase of one
of these items actually decreases the likelihood of purchasing the other. Without fully understanding this phenomenon, we could easily make unwise business decisions based on Rule (6.6).
2. From Association Analysis to Correlation Analysis
As we have seen so far, the support and confidence measures are insufficient at filtering out uninteresting association rules. To tackle this weakness, a correlation measure can be used to augment the support–confidence framework for association rules. This leads to correlation rules of the form
That is, a correlation rule is measured not only by its support and confidence but also by the correlation between itemsets A and B. There are many different correlation measures from which to choose. In this subsection, we study several correlation measures to determine which would be good for mining large data sets.
Lift is a simple correlation measure that is given as follows. The occurrence of itemset A is independent of the occurrence of itemset B if P(A U B) = P(A)P(B); otherwise, itemsets A and B are dependent and correlated as events. This definition can easily be extended to more than two itemsets. The lift between the occurrence of A and B can be measured by computing
If the resulting value of Eq. (6.8) is less than 1, then the occurrence of A is negatively correlated with the occurrence of B, meaning that the occurrence of one likely leads to the absence of the other one. If the resulting value is greater than 1, then A and B are positively correlated, meaning that the occurrence of one implies the occurrence of the other. If the resulting value is equal to 1, then A and B are independent and there is no correlation between them.
Correlation analysis using lift:
Correlation analysis using lift. To help filter out misleading “strong” associations of the form A=>B from the data of Example 6.7, we need to study how the two itemsets, A and B, are correlated. Let refer to the transactions of Example 6.7 that do not contain computer games, and
refer to those that do not contain videos. The transactions can be summarized in a contingency table, as shown in Table 6.6. From the table, we can see that the probability of purchasing a computer game is P({game}) = 0.60, the probability of purchasing a video is P({video}) = 0.75, and the probability of purchasing both is P({game, video}) = 0.40. By Eq. (6.8), the lift of Rule (6.6) is P({game, video})/P({game}).P({video}) = 0.40/(0.60X0.75) = 0.89. Because this value is less than 1, there is a negative correlation between the occurrence of {game} and {video}.
Correlation analysis using . To compute the correlation using
analysis for nominal data, we need the observed value and expected value (displayed in parenthesis) for each slot of the contingency table, as shown in Table 6.7. From the table, we can compute the
value as follows:
Multilevel Association Rule in data mining
Multilevel Association Rule :
Association rules created from mining information at different degrees of reflection are called various level or staggered association rules.
Multilevel association rules can be mined effectively utilizing idea progressions under a help certainty system.
Rules at a high idea level may add to good judgment while rules at a low idea level may not be valuable consistently.
Utilizing uniform least help for all levels :
- At the point when a uniform least help edge is utilized, the pursuit system is rearranged.
- The technique is likewise straightforward, in that clients are needed to indicate just a single least help edge.
- A similar least help edge is utilized when mining at each degree of deliberation. (for example for mining from “PC” down to “PC”). Both “PC” and “PC” discovered to be incessant, while “PC” isn’t.
Needs of Multidimensional Rule :
- Sometimes at the low data level, data does not show any significant pattern but there is useful information hiding behind it.
- The aim is to find the hidden information in or between levels of abstraction.
Approaches to multilevel association rule mining :
- Uniform Support(Using uniform minimum support for all level)
- Reduced Support (Using reduced minimum support at lower levels)
- Group-based Support(Using item or group based support)
Let’s discuss one by one.
- Uniform Support –
At the point when a uniform least help edge is used, the search methodology is simplified. The technique is likewise basic in that clients are needed to determine just a single least help threshold. An advancement technique can be adopted, based on the information that a progenitor is a superset of its descendant. the search keeps away from analyzing item sets containing anything that doesn’t have minimum support. The uniform support approach however has some difficulties. It is unlikely that items at lower levels of abstraction will occur as frequently as those at higher levels of abstraction. If the minimum support threshold is set too high it could miss several meaningful associations occurring at low abstraction levels. This provides the motivation for the following approach. - Reduce Support –
For mining various level relationship with diminished support, there are various elective hunt techniques as follows.
- Level-by-Level independence –
This is a full-broadness search, where no foundation information on regular item sets is utilized for pruning. Each hub is examined, regardless of whether its parent hub is discovered to be incessant. - Level – cross-separating by single thing –
A thing at the I level is inspected if and just if its parent hub at the (I-1) level is regular .all in all, we research a more explicit relationship from a more broad one. If a hub is frequent, its kids will be examined; otherwise, its descendant is pruned from the inquiry. - Level-cross separating by – K-itemset –
A-itemset at the I level is inspected if and just if it’s For mining various level relationship with diminished support, there are various elective hunt techniques. - Level-by-Level independence –
This is a full-broadness search, where no foundation information on regular item sets is utilized for pruning. Each hub is examined, regardless of whether its parent hub is discovered to be incessant. - Level – cross-separating by single thing –
A thing at the 1st level is inspected if and just if its parent hub at the (I-1) the level is regular .all in all, we research a more explicit relationship from a more broad one. If a hub is frequent, its kids will be examined otherwise, its descendant is pruned from the inquiry. - Level-cross separating by – K-item set –
A-item set at the I level is inspected if and just if its corresponding parents A item set (i-1) level is frequent.
- Group-based support –
The group-wise threshold value for support and confidence is input by the user or expert. The group is selected based on a product price or item set because often expert has insight as to which groups are more important than others.
Example –
For e.g. Experts are interested in purchase patterns of laptops or clothes in the non and electronic category. Therefore low support threshold is set for this group to give attention to these items’ purchase patterns.
Data Mining Multidimensional Association Rule
In Multi dimensional association rule Qualities can be absolute or quantitative.
- Quantitative characteristics are numeric and consolidates order.
- Numeric traits should be discretized.
- Multi dimensional affiliation rule comprises of more than one measurement.
- Example –buys(X, “IBM Laptop computer”)buys(X, “HP Inkjet Printer”)
Approaches in mining multi dimensional affiliation rules :
Three approaches in mining multi dimensional affiliation rules are as following.
-
Using static discretization of quantitative qualities :
- Discretization is static and happens preceding mining.
- Discretized ascribes are treated as unmitigated.
- Use apriori calculation to locate all k-regular predicate sets(this requires k or k+1 table outputs). Each subset of regular predicate set should be continuous.
Example –
If in an information block the 3D cuboid (age, pay, purchases) is continuous suggests (age, pay), (age, purchases), (pay, purchases) are likewise regular.Note –
Information blocks are appropriate for mining since they make mining quicker. The cells of an n-dimensional information cuboid relate to the predicate cells. -
Using powerful discretization of quantitative traits :
- Known as mining Quantitative Association Rules.
- Numeric properties are progressively discretized.
Example –:
age(X, "20..25") Λ income(X, "30K..41K")buys ( X, "Laptop Computer")
-
Grid FOR TUPLES :
Using distance based discretization with bunching –
This id dynamic discretization measure that considers the distance between information focuses. It includes a two stage mining measure as following.- Perform bunching to discover the time period included.
- Get affiliation rules via looking for gatherings of groups that happen together.
The resultant guidelines may fulfill –
- Bunches in the standard precursor are unequivocally connected with groups of rules in the subsequent.
- Bunches in the forerunner happen together.
- Bunches in the ensuing happen together.
What is Constraint-Based Association Mining?
A data mining procedure can uncover thousands of rules from a given set of information, most of which end up being independent or tedious to the users. Users have a best sense of which “direction” of mining can lead to interesting patterns and the “form” of the patterns or rules they can like to discover.
Therefore, a good heuristic is to have the users defines such intuition or expectations as constraints to constraint the search space. This strategy is called constraint-based mining.
Constraint-based algorithms need constraints to decrease the search area in the frequent itemset generation step (the association rule generating step is exact to that of exhaustive algorithms).
The general constraint is the support minimum threshold. If a constraint is uncontrolled, its inclusion in the mining phase can support significant reduction of the exploration space because of the definition of a boundary inside the search space lattice, following which exploration is not needed.
The important of constraints is well-defined − they create only association rules that are appealing to users. The method is quite trivial and the rules space is decreased whereby remaining methods satisfy the constraints.
Constraint-based clustering discover clusters that satisfy user-defined preferences or constraints. It depends on the characteristics of the constraints, constraint-based clustering can adopt rather than different approaches.
The constraints can include the following which are as follows –
Knowledge type constraints − These define the type of knowledge to be mined, including association or correlation.
Data constraints − These define the set of task-relevant information such as Dimension/level constraints − These defines the desired dimensions (or attributes) of the information, or methods of the concept hierarchies, to be utilized in mining.
Interestingness constraints − These defines thresholds on numerical measures of rule interestingness, including support, confidence, and correlation.
Rule constraints − These defines the form of rules to be mined. Such constraints can be defined as metarules (rule templates), as the maximum or minimum number of predicates that can appear in the rule antecedent or consequent, or as relationships between attributes, attribute values, and/or aggregates.
The following constraints can be described using a high-level declarative data mining query language and user interface. This form of constraint-based mining enables users to define the rules that they can like to uncover, thus by creating the data mining process more efficient.
Furthermore, a sophisticated mining query optimizer can be used to deed the constraints defined by the user, thereby creating the mining process more effective. Constraint-based mining boost interactive exploratory mining and analysis.
Classification Using Frequent Patterns in Data Mining
A data mining approach called frequent pattern mining is used to find recurring patterns in a dataset. It is a kind of unsupervised machine-learning technique that looks for and identifies patterns in data using algorithms. This method can be applied to find products that are frequently purchased together or to find products that are more likely to be purchased by particular demographic groups. Numerous applications of this method include client segmentation, fraud detection, and marketing analysis. Frequent pattern mining can be utilized in classification tasks to identify the patterns that are most likely related to a particular class.
Frequent patterns refer to item sets, subsequences, or substructures that appear frequently in a data set.
It works by scanning a data collection for common patterns, or item sets, and then utilizing those patterns to categorize previously undiscovered data items. Once learnt, the patterns may be utilized to categorize previously unknown data items, such as new consumer purchases or new customer behaviours. This categorization may be used for a number of purposes, including forecasting customer turnover and detecting fraudulent activity.