In this paper we present an adaptive collaborative filtering algorithm using Fish School Search. The proposed algorithm use not only rating information but also user demographic information and interests to improve similarity measurement. This algorithm adaptive to different user, where it could learn the best combination of features weight, leading to a better prediction. The experiment result shows that the proposed algorithm outperforms other collaborative filtering method. And on our knowledge, this is the first time Fish School Search applied in recommendation system domain.
|Number of pages||4|
|Journal||International Journal of Software Engineering and its Applications|
|Publication status||Published - 2014 Jan 1|
- Collaborative filtering
- Fish school search
- Recommendation systems
ASJC Scopus subject areas