User based Recommender Systems using Implicative Rating Measure

Abstract

Abstract-This paper proposes the implicative rating measure developed on the typicality measure. The paper also proposes a new recommendation model presenting the top N items to the active users. The proposed model is based on the user-based collaborative filtering approach using the implicative intensity measure to find the nearest neighbors of the active users, and the proposed measure to predict users' ratings for items. The model is evaluated on two datasets MovieLens and CourseRegistration, and compared to some existing models such as: the item based collaborative filtering model using the Jaccard measure, the user based collaborative filtering model using Jaccard measure, the popular items based model, the latent factor based model, and the association rule based model using the confidence measure. The experimental results show that the performance of the proposed model is better when compared to other five models.

A conversational recommender system for diagnosis using fuzzy rules

“…This information allows the optimization of the selection strategy of key-terms through user feedback. Other filtering tecnhiques based on statistical measures are used in (Phan et al, 2017). Specifically, in this work, we focus on Conversational Recommender systems (Christakopoulou et al, 2016), which work by interacting with the user and building a conversation that ends in the recommendation.…”