A webpage similarity measure for web sessions clustering using sequence alignment

Web sessions clustering is a process of web usage mining task that aims to group web sessions with similar trends and usage patterns into clusters. This process is crucial for effective website management, web personalization and developing web recommender systems. Accurate clustering of web sessions is highly dependent to the similarity measure defined to compare web sessions. In this paper, we propose a similarity measure for comparing web sessions. The sequential order of web navigations in sessions is considered using sequence alignment method. Furthermore, we propose to consider the usage similarity of two web sessions based on the time a user spends on a webpage, and also the frequency of visit of each page within the session. The proposed method is validated by clustering a collection of web sessions using an agglomerative clustering technique and comparing the results with available methods. The experimental results show effectiveness of the proposed method to capture the properties of web session data.