A Study of Collaborative Filtering Approach for Temporal Dynamic Web Data

Collaborative filtering is widely used and popular tool these days. In collaborative filtering, user preference data, collected over a long period of time, is exploited to predict interest on the unseen items the basis of users with similarity interests. The similarity amongst the items is determined by the similarity function as weighted average of the ratings given by the users. In this paper, an improved similarity function for collaborative filtering is proposed that incorporates the time when the item was rated. This allows the collaborative filtering to capture the data more accurately and efficiently.