Group Recommendations Based on Hesitant Fuzzy Sets

Publication Type:
Journal Article
International Journal of Intelligent Systems, 2018, 33 (10), pp. 2058 - 2077
Issue Date:
Filename Description Size
Castro_et_al-2018-International_Journal_of_Intelligent_Systems.pdfPublished Version889.72 kB
Adobe PDF
Full metadata record
© 2017 Wiley Periodicals, Inc. Group recommender systems (GRSs) recommend items that are used by groups of people because certain activities, such as listening to music, watching a movie, dining in a restaurant, etc., are social events performed by groups of people sharing their tastes, and their choices affect all of them. GRSs help groups of people making choices in overloaded search spaces according to all group members preferences. A common GRS scheme aggregates users preferences to generate a group preference profile. However, the aggregation process may imply a loss of information, negatively affecting different properties of the GRS such as diversity of group recommendations, which is an important quality factor because of such recommendations are targeted to groups formed by users with individual and possibly conflicting preferences. To avoid and manage the loss of information caused by aggregation, this paper proposes to keep all group members preferences by using hesitant fuzzy sets (HFSs) and interpreting such information like the group hesitation about their preferences that will be used in the group recommendation process. To evaluate the performance and rank quality of the HFS GRS proposal, a case study is carried out.
Please use this identifier to cite or link to this item: