Exploring Fuzzy Rating Regularities for Managing Natural Noise in Collaborative Recommendation

DOI

10.2991/ijcis.d.191115.001How to use a DOI?

Keywords

Recommender systems; Natural noise; Regularities; Fuzzy logic

Abstract

Recommender systems have played a relevant role in e-commerce for supporting online users to obtain suggestions about products that best fit their preferences and needs in overloaded search spaces. In such a context, several authors have proposed methods focused on removing the users' inconsistencies when they rate items, so-called natural noise, improving in this way the recommendation performance. The current paper explores the use of rating regularities for managing the natural noise in collaborative filtering recommendation, having as key feature the use of fuzzy techniques for coping with the uncertainty associated to such scenarios. Specifically, such regularities are used for representing common rating patterns and thus detect noisy ratings when they tend to contradict such patterns. An experimental study is developed for showing the performance of the proposal, as well as analyzing its behavior in contrast to previous natural noise management procedures.

Copyright

© 2019 The Authors. Published by Atlantis Press SARL.

Open Access

This is an open access article distributed under the CC BY-NC 4.0 license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)
View full text (HTML)

TY  - JOUR
AU  - Raciel Yera
AU  - Manuel J. Barranco
AU  - Ahmad A. Alzahrani
AU  - Luis Martínez
PY  - 2019
DA  - 2019/11/25
TI  - Exploring Fuzzy Rating Regularities for Managing Natural Noise in Collaborative Recommendation
JO  - International Journal of Computational Intelligence Systems
SP  - 1382
EP  - 1392
VL  - 12
IS  - 2
SN  - 1875-6883
UR  - https://doi.org/10.2991/ijcis.d.191115.001
DO  - 10.2991/ijcis.d.191115.001
ID  - Yera2019
ER  -