2025

Conférence

21st International Conference on Artificial Intelligence Applications and Innovations (AIAI 2025) : 381-391

Recommendation systems have become essential for filtering the vast amounts of information available on the Internet. Traditional collaborative filtering methods face challenges such as data sparsity and scalability issues. To address these limitations, we propose ColBic, a novel collaborative filtering approach based on biclustering and Iterative Local Search (ILS). Our method improves the accuracy of the recommendation by grouping users and items into dense biclusters and refining them through iterative optimization. Experimental results on the MovieLens-100K and MovieLens-1M datasets demonstrate that ColBic outperforms traditional collaborative filtering methods in terms of accuracy and coverage.

@InProceedings{10.1007/978-3-031-96228-8_28,
author="Achouri, Eya
and Ayadi, Wassim",
editor="Maglogiannis, Ilias
and Iliadis, Lazaros
and Andreou, Andreas
and Papaleonidas, Antonios",
title="ColBic: A New Biclustering-Based Collaborative Filtering",
booktitle="Artificial Intelligence Applications and Innovations",
year="2025",
publisher="Springer Nature Switzerland",
address="Cham",
pages="381--391",
abstract="Recommendation systems have become essential for filtering the vast amounts of information available on the Internet. Traditional collaborative filtering methods face challenges such as data sparsity and scalability issues. To address these limitations, we propose ColBic, a novel collaborative filtering approach based on biclustering and Iterative Local Search (ILS). Our method improves the accuracy of the recommendation by grouping users and items into dense biclusters and refining them through iterative optimization. Experimental results on the MovieLens-100K and MovieLens-1M datasets demonstrate that ColBic outperforms traditional collaborative filtering methods in terms of accuracy and coverage.",
isbn="978-3-031-96228-8"
}