A Recommender System for Infrequent Purchased Products based on User Navigation and Product Review Data

Recommender Systems (RS) help users to make decisions about which product to purchase from the vast amount of products available on the Internet. Currently, many of the existing recommender systems are developed for recommending frequently purchased products where a large amount of explicit ratings data is available to predict user preferences. However, it is difficult to collect this data for products that are infrequently purchased by the users, and, thus, user profiling becomes a major challenge for recommending such products. This paper proposes a recommender system approach that exploits user navigation and product review data for generating user and product profiles, which are used for recommending infrequently purchased products. The evaluation result shows that the proposed approach, named Adaptive Collaborative Filtering (ACF), which utilizes user and product profiles, outperforms the Query Expansion (QE) approach that only utilizes product profiles to recommend products. ACF also performs better than Basic Search (BS) approach, which is widely applied by the current e-commerce applications.