Collaborative Filtering Recommendation System Application based on Stereotype Model

초록

In this paper, we present an improved method to slove the sparsity problems of collaborative filtering by defining a user model as a stereotype using demographic information based on user's characteristics and adopt stereotype information to make up for similiarity correction. The similarity between users is only determined by the ratings given to co-rated items that have not been rated by both users are ignored. To solve this problem, we add virtual neighbor's rating using demographic information of neighbors for improving prediction accuracy. We show improved accuracy by comparing between the traditional Peason Correlation coefficient and the proposed method.