Group Recommender Systems-Evolutionary Approach Based on Consensus with Ties

摘要

The issue regarding aggregation of multiple rankings into one consensus ranking is an interesting research subject in a ubiquitous scenario that includes a group of users. For minimizing the fitness value of Kendall tau distance (KtD), the well-known optimal aggregation method of Kemeny is used to generate an aggregated list from the input lists. A primary goal of our work is to recommend a list of items or permutation that can effectively handle the problem of full ranking with ties using consensus (FRWT-WC). Additionally, in real applications, most of the studies have focused on without ties. However, the rankings to be aggregated may not be permutations where elements have multiple choices ordered set, but they may have ties where some elements are placed at the same position. In this work, in order to handle problem of FRWT in GRS using consensus measure function, KtD are used as fitness function. Experimental result are presents that our proposed GRS based on Consensus for FRWT (GRS-FRWT-WC) outperforms well-knows baseline GRS techniques. In this work, we design and evolve an innovative method to solve the problem of ties in GRS based on consensus and results show that efficiency of group does not certainly reduce in which the group has similar-minded user.