Arash Beheshtian-Ardakani - Department of Industrial Engineering Iran University of Science and Technology Tehran, Iran
Mohammad Fathian - Department of Industrial Engineering Iran University of Science and Technology Tehran, Iran
Babak Nouri-Moghaddam - Department of Industrial Engineering Iran University of Science and Technology Tehran, Iran
Yashar Gorgani - Department of Electrical and Computer Engineering Graduate University of Advanced Technology Kerman, Iran

Electronic retailers are the firms that receive the products from manufacturers and sell them directly to consumers. The determination of the products that are more likely to be purchased by the customers is important for electronic retailers. In this study, we have investigated the customers’ buying patterns in an e-commerce website by using data mining algorithms. The data of the customers’ orders have been collected and processed from the website of an electronic retailing company, and then the customer loyalty is calculated. Based on the ordered items, the customers have been clustered and the important properties of the orders are identified by using Forest Optimization Algorithm (FOA). Then the buying patterns of each cluster are extracted by using the association rules. To predict the customer behavior in the future, we create a decision tree for each buying pattern to specify the conditions in which the rules are occurred. Finally, via direct marketing and based on the discovered models, the information of the product bundles will be sent to customers’ email.

Direct marketing, Forest Optimization Algorithm, Market segmentation, Feature selection