Performance Improvements and Efficient Approach for Mining Periodic Sequential Access Patterns

Surfing the Web has become an important daily activity for many users. Discovering and understanding web users’ surfing behavior are essential for the development of successful web monitoring and recommendation systems. To capture users’ web access behavior, one promising approach is web usage mining which discovers interesting and frequent user access patterns from web usage logs. Web usage mining discovers interesting and frequent user access patterns from web logs. Most of the previous works have focused on mining common sequential access patterns of web access events that occurred within the entire duration of all web access transactions. However, many useful sequential access patterns occur frequently only during a particular periodic time interval due to user browsing behaviors and habits. It is therefore important to mine periodic sequential access patterns with periodic time constraints. In this paper, we propose an efficient approach, known as TCSMA (Temporal Conditional Sequence Mining Algorithm), for mining periodic sequential access patterns based on calendar-based periodic time constraint. The calendar-based periodic time constraints are used for describing real-life periodic time concepts such as the morning of every weekend. The mined periodic sequential access patterns can be used for temporal-based personalized web recommendations. The performance of the proposed TCSMA is evaluated and compared with a modified version of Web Access Pattern Mine for mining periodic sequential access patterns.