Simple time-aware and social-aware user similarity for a KNN-based recommender system

Authors:
Andrés Moreno;Harold Castro;Michel Riveill
Affiliations:
University of los Andes, Bogotá, Colombia;University of los Andes, Bogotá, Colombia;Laboratoire I3S (Université de Nice Sophia Antipolis), Sophia Antipolis, France
Venue:
Proceedings of the 2nd Challenge on Context-Aware Movie Recommendation
Year:
2011

Citing 4
Cited 1

GroupLens: an open architecture for collaborative filtering of netnews

CSCW '94 Proceedings of the 1994 ACM conference on Computer supported cooperative work
Utilizing Physical and Social Context to Improve Recommender Systems

WI-IATW '07 Proceedings of the 2007 IEEE/WIC/ACM International Conferences on Web Intelligence and Intelligent Agent Technology - Workshops
Collaborative filtering with temporal dynamics

Proceedings of the 15th ACM SIGKDD international conference on Knowledge discovery and data mining
Simple time-biased KNN-based recommendations

Proceedings of the Workshop on Context-Aware Movie Recommendation

Group recommendation in context

Proceedings of the 2nd Challenge on Context-Aware Movie Recommendation

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper we report the results of the experiment carried out by our institutions for participating in the track 1 of the the 2011 CAMRa challenge workshop. We ran some variations of a traditional user-based neighborhood recommender system based on two simple ideas: (1) Force the inclusion of household members into the neighborhood of the user and (2) increase the similarity of users that use the system if they use the system at similar time slots. The approaches are evaluated using the MAP, P@5, P@10 and AUC metrics. Results show that a small improvement is achieved on of the chosen metrics when comparing the social and time strategies to a traditional KNN approach.