Evaluation of Recommended Web Pages Obtained using a Novel Approach
With the increasing number of web users, web-based organizations are keen to analyze the on-line browsing behavior of the users in their web site and identify their interest immediately in a session. The analysis of the user's current interest based on the navigational behavior may help the organizations to guide them in their browsing activity and provide recommendations in a shorter span of time. There are various approaches for recommendations but the selection of the recommendation approach is determined by the evaluation measures. This paper focuses on the evaluation measures of the recommendation systems to determine their "Correctness" and "Completeness". The experimental results performed on real usage data from a commercial web site show a significant improvement in the recommendation effectiveness of the proposed system.