Precedence Mining in Group Recommender Systems

Abstract

Abstract. We extend the Precedence mining model for personal recommendation as outlined in Parameswaran et.al.,[6] in three different ways. Firstly, we show how precedence mining model can be used for recommending items of interest to a group of users (group recommendation) and compare and contrast our model with traditional group recommendation models like collaborative and Hybrid. Secondly, we extend the precedence mining model to incorporate ratings for items and experimental results show that the goodness of recommendation is improved. The third extension is related to the issue of new items being ignored which is a fundamental problem plaguing collaborative and precedence mining algorithms. When recommendations are based on other users interests (like in Collaborative recommender systems) the possibility of not recommending a new item which has not been consumed by many of the users is high though the new item may be of interest to the target user. We outline two models, Vector precedence-mining and Hybrid precedence-mining that addresses this issue.

Untitled

“…Merging strategy and Virtual User strategy are two conventional approaches for grouping. According to Kagita et al [26], there are three ways to implement the Merging strategy: merged profiles, merging recommendation, and merging score. Incidentally, Kagita et al [27] also illuminated the Virtual User strategy.…”

Virtual user approach for group recommender systems using precedence relations

“…It is to emphasize here that there has not been any attempt of constructing virtual user profile in this context. In this paper we examine the characteristics of our scheme of virtual user strategy, which we introduced in preliminary versions in [11,12]. The new scheme of virtual user strategy is described and we examine its properties and show that our new scheme is more general than the earlier known strategy which is based on common items.…”

A Novel Social-Choice Strategy for Group Modeling in Recommender Systems

“…Merge profiles approach creates the recommendation list from merging the profiles of each group member which in turn is based on individual ratings for content or genre. For instance, consider two user group [15], [16], [17]. In the merging recommendation approach, GRS first generates a recommendation for each user individually based on personalized profiles.…”