Data warehouse can be defined as the integrated, subject oriented and non volatile collection of data. The data collected in warehouse help analysts in organized decision making. The data stored can undergo frequent changes on daily basis. The warehouse data will provide an overview about a product which will further assist in improving the business. The consolidated data from data warehouses are presented in multidimensional view. The multidimensional analysis of warehouse data leads to data generalization.
Importants points to remember about Data Warehouse
- Data warehouse is a database which is separated from operational database of an organization.
- Operational database require frequent update while no frequent database updates are done in data warehouse.
- All the strategic decisions in an organization are done using data warehouse as a source.
- Application systems can be integrated and diversified using data warehouse.
- Historical data analysis are done using data warehouse.
Related: What is Index in SQL?
Features of Data warehouse
- Subject Oriented – Data warehouse is subject oriented because it provide information about subjects rather than providing information about organization.
- Integrated – Data warehouses are integrated because they are constructed by integrating data from relational databases and flat files.
- Time Variant – The data stored in data warehouse can be identified using a particular time period in which the data was stored.
- Non-Volatile – Data stored in data warehouse will not be erased when new data is added.
Types of Data Warehouses
Data warehouses are classified into three types
- Information Processing
- Analytical Processing
- Data Mining
Data mining is a process in which raw and unprocessed data are converted into useful information. Using data mining companies will understand the interests of customers. Almost all marketing strategies of companies are based on the data mining. The effective results of data mining depends upon data warehousing, data collection and data processing.
The loyalty cards offered by many shops are good example for data mining. The regular customers use these loyalty card for considerable amount of discount. The companies use the data collected through loyalty card to access which type of goods have more demand and the kind of goods purchased by regular customers. Using data mining the organizations can be alerted if they are losing customers. Data mining is also used in detecting credit frauds because all the transaction details of customers are saved using data mining.
Various fields in which data mining is used
- Customer Profiling
- Identifying customer requirements
- Cross Market Analysis
- Target Marketing
- Determining Purchasing Pattern
- Providing Information
The following are the themes used in data mining
- Data Reduction
- Data Compression
- Pattern Discovery
- Inductive databases
- Microeconomic View
Read: More Interview Questions
Difference between Data Mining and Data Warehousing
- Data mining is based on pattern recognition logic which trend against larger data pool.
- Data warehouse extracts and stores data for future references.
- Data mining is a range of business process while data warehouse is like typical software package which can process some data.
- Data mining is used in various marketing programs while data warehousing provide the data for designing the logic of data mining.
- Data warehouse is designed to be a software product but data mining is a logic design based on the results provided by data warehouse.
- The entire business process is based on the query of data warehousing.