Sem Spirit

Association Rule Learning

The search for association rules is a popular method studied in depth in the field of data mining. The goal is to discover relationships of interest to the statistician between two or more variables stored in very large databases using association rule search algorithms. Based on the concept of strong relationships, these algorithms present association rules whose purpose is to discover similarities between products in data captured on a large scale in the IT systems of retail outlets of supermarket chains. For example, a rule found in supermarket sales data might indicate that a customer buying tomato sauce and beef simultaneously would be likely to buy pasta. Such information can be used as a basis for making marketing decisions such as promotions or well-chosen locations for related products. In addition to the above examples of the housewife basket, the rules of association are used today in several areas including the web search, intrusion detection and bioinformatics.