Designing a Well-Structured E-Shop Using Association Rule Mining
Asem Omari, Stefan Conrad, Sadet Alcic · 2007
Many commercial companies collect large quantities of data from daily operations. For example, customer orders or purchase data are collected daily at the counters of grocery stores. Data mining is applied on such kind of data to extract patterns that could be useful to learn about the purchasing behavior of the customers. Such information are used to support a variety of business related tasks. For example, the investment of that kind of information in building a Website for a grocery store. Association rule mining is one of the techniques used to mine databases. Association rule mining is the discovery of association rules showing attribute values that occur frequently together. In this paper, we have discovered association rules from a grocery store dataset which represents customer transactions in that grocery store. Those rules have been invested to design a well- structured Website prototype for that grocery store. Promising results, that could affect the process of Website design, have been found. The experiments showed that our method can reduce the cost up to 90% in some transactions.