A model‐based approach to user preference discovery in multi‐criteria recommender system using genetic programming

Abstract

Multi-criteria recommender systems (MCRSs) provide suggestions to users based on their preferences to various criteria. Incorporation of criteria ratings into recommendation framework can provide quality recommendations to users because these ratings can elicit users' preferences efficiently. However, elicitation of user's overall preference based on criteria ratings is a key issue in MCRS. Even though several aggregation methods for the elicitation of users' overall preference have been investigated in the literature, no method has been shown the superiority under all circumstances. Therefore, we propose a model based approach to user preference discovery in multi-criteria RS using genetic programming (GP). In this work, we suggest three-stage process to generate recommendations to users. First, we learn user preference transformation function to aggregate criteria ratings by using GP, and then we utilize the preference function, so derived, for computing similarities in MCRS. Finally, items are recommended to users.Experimental results on Yahoo! Movies dataset show the superiority of our proposed approach in comparison to other aggregation approaches.

Adaptive genetic algorithm for user preference discovery in multi-criteria recommender systems

“…Similar to the CF technique, the MCRS has been classified as memory-based and model-based approaches [ 2 ]. The model-based approach involves developing a system that identifies a function that determines how single criteria and multi-criteria ratings relate to one another [ 14 ]. On the other hand, the memory-based approach expands traditional RS by identifying user similarities based on different criteria and aggregating these similarities using different methods.…”

“…Results showed that the GA-based approaches gave better performance and the multi-heuristic GA-based approach outperformed other approaches. On the other hand, GP was employed in MCRS to learn criteria weights [ 14 , 19 ]. Where GP was used to learn a function to transform user preferences into an aggregate of criteria ratings.…”

“…Similarly, heuristic approaches also aggregate the user's preferences based on various criteria. Aggregating criteria ratings is a crucial step, and if not performed properly, it may lower the value of having multiple factors of an item [ 14 ]. Therefore, to develop a truly personalized MCRS, these criteria need to be efficiently incorporated and their individual weights must be learned and adjusted for actual user preference discovery in MCRS.…”