Attribute Clustering Based Collaborative Filtering
Liu Mei Zhang, Jian Feng, Di Lu, Yi Chuan Wang · Applied Mechanics and Materials · 2014
The paper proposed an attribute clustering based collaborative filtering algorithm for recommendation. It utilizes similarity to filter out redundant attributes by feature selection. Then by incorporating K-Means clustering, it is able to effectively solve the rating scale problems existing in the traditional collaborative filtering recommendation algorithm. The algorithm is verified by real data sets. Experiments use location information for clustering the restaurant data. By integration of users rating on restaurant service and external impression the experiment study combined the collaborative filtering philosophy to provide recommendation service for users. Experimental results show that compared with the item rating based recommended algorithm, the algorithm has ideal recommended quality and improved accuracy, and then it has reduced the data sparsity.