A Particle Swarm Approach to Collaborative Filtering based Recommender Systems through Fuzzy Features

Collaborative filtering (CF) either memory based or model based, has been emerged as an information filtering tool that provides effective recommendations to users utilizing the experiences and opinions of their similar neighbors when they interact with large information spaces. Memory based CF is more accurate than model based CF but it is less scalable. Our work in this paper is an attempt towards introducing a recommendation strategy (FPSO-CF) based on user hybrid features that retains the accuracy of memory – based CF as well as the scalability of model-based CF in an efficient manner. Since most user features are imprecise in nature, therefore these can be represented more naturally by using fuzzy sets. In this work, we employ particle swarm optimization algorithm (PSO) to learn user weights on various features and use fuzzy sets for representing user features efficiently. Effectiveness of our proposed RS (FPSO-CF) is demonstrated through experimental results in terms of various performance measures using the MovieLens dataset.