A Hybrid Information Filtering Model
Xun Wang, Yi Xie, Biwei Li · 2006
To address the issues that user evaluation data is extremely sparse, the user-accessing matrix based on Web log mining is established, which takes the frequencies of user accessing, browsing time and the length of the pages into consideration. Furthermore, a novel collaborative filtering algorithm based on Web page rating prediction is proposed. This method predicts Web page ratings that users have not rated by the similarity of Web page, and uses the correlative similarity measure to find the target users' neighbors. Eventually, a hybrid-filtering model is proposed to overcome the drawbacks of the content-based filtering and the collaborative filtering models. The experimental results show that the hybrid-filtering model can efficiently cope with the faults of traditional filtering models and greatly improve the recommendation quality