What is Data Warehouse Mining?

Data warehouse mining is the analysis of information contained in one or more databases in order to make the information useful. These databases, or data warehouses, are a central depository for data. Companies aggregate the information they collect on their customers in a data warehouse. Once the information has been collected, it is "mined," and useful information is extracted from it to produce information that can help the company make business decisions that will increase profits or reduce costs. Retailers frequently use data warehouse mining to analyze and predict the behavior of their customers.

For example, when a shopper goes to the supermarket and gives the cashier her frequent shopper card, information about her purchases is collected and stored in the company's data warehouse. A supermarket chain will have millions of pieces of data on what people buy, when, in what quantities, and at what price. A store may know that 50,000 packages of frozen peas were sold last year, but that information alone is not particularly helpful. If the data warehouse mining reveals, however, that 75% of those frozen peas were sold during months when fresh peas were not available, or that 10% of the peas were sold in the two weeks leading up to Thanksgiving, the company may be able to use that information to increase their annual sales of frozen peas.

Companies can employ data warehouse mining techniques to predict future sales. Data mining can also help them to estimate the impact of stocking and pricing decisions. At the supermarket, data mining might keep the stores from running out of frozen peas in the event of a poor crop of fresh peas in a given year.

Data mining regression is a data mining technique that is used to show what is likely to happen to a data value if something in the equation is changed. Using the supermarket example, regression would predict the level of frozen pea sales if fresh peas increased in price. Regression uses historical data and applies a formula to it, which predicts future behavior.

Companies will often use a data warehouse mining software application to collect and mine their data. The correct application is determined by the amount of data they have and the type of analysis they want to do. Choosing the correct data mining tool is critical to gathering and interpreting useful data.