Predicting Customers’ Behavior Using Web-Content Mining and Web-Usage Mining

Today, e-commerce has become a competitive space for online retailers. Therefore, personalization has become a vital part of e-commerce websites’ success, challenging marketers and researchers. This study aims to provide a model for web personalization and mining user interests using a hybrid web usage and web content mining approach. The navigational patterns of web users and the interests of each user on web pages of a Persian website were extracted through web usage mining and topic modeling. Users were then clustered using the dependency distribution algorithm, and 25 categories were extracted. To better understand the behavioral patterns of web users they were categorized using the Support Vector Machine algorithm based on the users’ interests and navigational behaviors. The most important result of the proposed system is that the patterns of users’ navigation are understandable, and the subsequent analyses will be much simpler.