Collabrative Filtering Recommendation Based on Users' Feature and Users' Drifting Interest
HE Xi-chun · Computer Knowledge and Technology · 2008
Collaborative filtering (CF) is the most successful recommended system to date,but traditional CF algorithm does not considerthe problem of drifting users' interests and users' feature which often results in poor recommendation. This paper combined two factors ofdrifting users' interests and user' feature. A new method of computing similarities between different users is developed which can enhanceaccurate of commendation.