A clustering based approach to improving the efficiency of collaborative filtering recommendation

• Developing an efficient collaborative filtering recommender system for E-commerce.
• Applying a self-constructing clustering algorithm to reduce the dimensionality related to the products.
• Reducing the huge correlation graph to a much smaller graph for faster computation.
• Demonstrating that efficiency is greatly Improved without degradation of the recommendation quality.

In collaborative filtering recommender systems, products are regarded as features and users are requested to provide ratings to the products they have purchased. By learning from the ratings, such a recommender system can recommend interesting products to users. However, there are usually quite a lot of products involved in E-commerce and it would be very inefficient if every product needs to be considered before making recommendations. We propose a novel approach which applies a self-constructing clustering algorithm to reduce the dimensionality related to the number of products. Similar products are grouped in the same cluster and dissimilar products are dispatched in different clusters. Recommendation work is then done with the resulting clusters. Finally, re-transformation is performed and a ranked list of recommended products is offered to each user. With the proposed approach, the processing time for making recommendations is much reduced. Experimental results show that the efficiency of the recommender system can be greatly improved without compromising the recommendation quality.

Overview of our approach.Figure optionsDownload as PowerPoint slide