Improving Collaborative Filtering Recommender System Results and Performance using Combination of Fuzzy Grey Wolf Optimizer Algorithm and Lion Optimization Algorithm

Abstract

Nowadays, recommender systems have reshaped the ways of information filtering between websites and the users in order to identify the users’ interests and generate product suggestions for the active users. Recommender systems are generally divided into three groups: Content-based, Knowledge-based, and collaborative-based, and in some cases hybrid. The main idea of collaborative filtering is that they predict a user’s interest in new items based on the recommendations of other people with similar interests. This Approach does not require having knowledge about items. Collaborative filtering has two main types: Memory-based and Model-based. Memory based Collaborative filtering makes use of user rating dataset to compute similarity index between set of users or set of items. The main purpose of this article is to offer a Memory-based Collaborative recommender system in order to optimize the results of Collaborative filtering algorithm. In the proposed method, the combination of fuzzy Grey Wolf Optimizer algorithm and Lion Optimization Algorithm is used to find the most similar users to the target user. The results of the proposed method confirmed a significant increment in Precision, Recall and F-measure in comparison with baseline methods.

Keywords

• recommender systems
• collaborative filtering
• metaheuristic algorithms
• grey wolf optimizer algorithm
• lion optimization algorithm