A Recommender System Based on Hesitant Fuzzy Linguistic Information with MAPPACC Approach

Recommender systems can make contributions to enterprises by meeting the demands of users and improving their satisfaction. However, because of the uncertainty and complexity of users’ preferences, the classical techniques are insufficient to sort out the suitable recommendations. Scholars have made progress to address uncertainty in recommender systems, but the existing studies neglected the uncertain linguistic information and failed to use them to efficiently provide personalized recommendations for individuals. Therefore, this paper demonstrates an item-based recommendation program combined with the Hesitant Fuzzy Linguistic Multi-criteria Analysis of Preferences by means of Pairwise Actions and Criterion Comparisons (HFL-MAPPACC) method to analyze hesitant fuzzy linguistic information. To provide personalized recommendations, preference degrees of the used items and tendency of evaluations are considered in the construction of this algorithm. Then, the proposed approach is implemented in a doctor recommender system to show its applicability. Ultimately, the validity of the proposed method and its superiority are discussed by comparative analyses.