Mitigation of the cold start problem in movie recommender systems using fuzzification of users’ social network information

Recommender systems can significantly assist users in making decisions by helping them find potentially relevant items from a vast range of choices. One significant challenge in the field of recommender systems is the "cold start problem," which arises when there is no prior information available for new users or items, and consequently, the system cannot make proper inferences. To address this issue, this study utilizes a multi-source dataset called the CSP Dataset, which includes user profiles, movie features, ratings, and contextual data. using this behavioral user dataset resolves the lack of prior information problem. additionally, a fuzzy membership function is employed to introduce fuzziness to the selected features of the dataset based on numerical variables, creating a fuzzy ratings measure for each combination of user, item, and context. These fuzzy ratings were then used to develop a fuzzy-based recommender system capable of handling user cold start scenarios. The findings indicate that the proposed approach outperforms the state-of-the-art methods as well as the existing methods in terms of accuracy.