Data mining refers to the process of searching for valuable business information in a large database, data warehouse, or data mart.

Data mining can perform two basic operations: (1) predicting trends and behaviors, and (2) identifying previously unknown patterns.

BI applications typically provide users with a view of what has happened; data mining helps to explain why it is happening, and it predicts what will happen in the future.

Regarding the first operation, data mining automates the process of finding predictive information in large databases.

Data mining can use data from past promotional mailings to identify those prospects who are most likely to respond favorably to future mailings. Another business problem that uses predictive information is the forecasting of bankruptcy and other forms of default.

Data mining can also identify previously hidden patterns in a single step.

Numerous data mining applications are used in business and in other fields. According to a Gartner report (www.gartner.com), most Fortune 1000 companies worldwide currently use data mining, as the following representative examples illustrate. Note that in most cases the purpose of data mining is to identify a business opportunity to create a sustainable competitive advantage.

Retailing and sales: Predicting sales, preventing theft and fraud, and determining correct inventory levels and distribution schedules among outlets. For example, retailers such as AAFES (stores on military bases) use Fraud Watch from SAP (www.sap.com) to combat fraud by employees in their 1,400 stores.

Banking: Forecasting levels of bad loans and fraudulent credit card use, predicting credit card spending by new customers, and determining which kinds of customers will best respond to (and qualify for) new loan offers.

Manufacturing and production: Predicting machinery failures, and finding key factors that help optimize manufacturing capacity.

Insurance: Forecasting claim amounts and medical coverage costs, classifying the most important elements that affect medical coverage, and predicting which customers will buy new insurance policies.

Policework: Tracking crime patterns, locations, and criminal behavior; identifying attributes to assist in solving criminal cases.

Healthcare: Correlating demographics of patients with critical illnesses, and developing better insights on how to identify and treat symptoms and their causes.

Marketing: Classifying customer demographics that can be used to predict which customers will respond to a mailing or buy a particular product.