Improve Quality of Recommendation System Using Hybrid Filtering Approach

DOI

10.2991/aer.k.211129.023How to use a DOI?

Keywords

recommender system; hybrid filtering; matrix factorization; collaborative filtering; content based filtering

Abstract

Recommendation systems are widely used on website platforms such as e-commerce, marketplaces, streaming movies to produce appropriate item recommendations for each user. The traditional memory-based collaborative filtering approach is currently used in recommender systems. This approach relies on users’ item rating as a basic approach for calculates the similarity of users’ responses about products to predict item recommendations, but the weakness is high prediction errors. This study aimed to reduce prediction errors from a memory-based collaborative filtering approach using hybrid filtering, so the recommendation system’s quality can be improved. The hybrid filtering approach combined a collaborative filtering approach based on a matrix factorization model and content-based filtering, which can reduce prediction errors to produce accurate item recommendations. The proposed method has been evaluated ten times using the root mean squared error to measure the prediction error. As a result, the hybrid filtering approach produced the smallest prediction error of 0.68, while memory-based collaborative filtering was 2.98. Based on the results, the hybrid filtering approach is better than memory-based collaborative filtering.

Copyright

© 2021 The Authors. Published by Atlantis Press International B.V.

Open Access

This is an open access article under the CC BY-NC license.

Download article (PDF)

TY  - CONF
AU  - Annas Al Amin
AU  - Andi Sunyoto
AU  - Hanif Al Fatta
PY  - 2021
DA  - 2021/11/30
TI  - Improve Quality of Recommendation System Using Hybrid Filtering Approach
BT  - Proceedings of the International Conference on Innovation in Science and Technology (ICIST 2020)
PB  - Atlantis Press
SP  - 103
EP  - 106
SN  - 2352-5401
UR  - https://doi.org/10.2991/aer.k.211129.023
DO  - 10.2991/aer.k.211129.023
ID  - AlAmin2021
ER  -