Deriving a Categorical Vector Space Model for Web Page Recommendations Based on Wikipedia’s Content

摘要

This work models web pages and users for web page recommendation by deriving a categorical vector space model (WikiVSM). We augment Wikipedia’s categorization system with keywords extracted from Wikipedia’ pages. By applying keyword matching, any web page can be mapped into WikiVSM. We compare WikiVSM’s performance with Vector Space Model (VSM) regarding recommending topically relevant web pages to individual users based on their browsing history. Results indicate that WikiVSM performs significantly better in generating recommendations. This is possibly due to features of Wikipedia and our augmentation.