Associating Expertized Information to Alleviate Sparsity Problem in Personalization
Ming-yu Lee, Chiung-Wei Huang, Hahn-Ming Lee · 2006
Personalization is an important technique in e-commerce. In this paper, we propose an approach to alleviate the sparsity and cold-start problems in the recommendation system for personalization. The expertized hierarchical classification information in library science is introduced and associated to enhance the similarity computation between books in our case. The enhanced similarities and preference ratings are used to estimate the missing values of preference rating table. Then by applying feature augmentation hybridization technique, the item-based collaborative filtering approach makes recommendations for users. To prove the performance, our evaluation is conducted offline on existing data set. From experimental results, the proposed recommendation system outperforms the classic item-based collaborative filtering approach in both recommendation quantities and qualities