Abstract

Current recommender systems struggle with recommending relevant information from an overwhelming body of data. As users interact with more and more items, it gets increasingly difficult to recognize which items are relevant. In this paper we present the concept of inferred contextual user profiles (CUPs) which extends the traditional user profile definition by describing the user in a given situation, or context. The approach is evaluated in the scope of the inferred profiles. In our evaluation, we infer two CUPs for each user, and use only one of the profiles, instead of the full user profile for recommending movies. We evaluate the model on a data snapshot from the Moviepilot movie recommendation website, with results showing a substantial improvement in terms of precision, recall and mean average precision.

@inproceedings{said2011c,
author = {Alan Said and Ernesto W. De Luca and Sahin Albayrak},
title = {Inferring Contextual User Profiles – Improving Recommender Performance},
booktitle = {Proceedings of the 3rd RecSys Workshop on Context-Aware Recommender Systems},
year = {2011},
location = {Chicago, IL, USA},
}