A switching multi-level method for the long tail recommendation problem

Alshammari, Gharbi, Jorro-Aragoneses, Jose L., Polatidis, Nikolaos, Kapetanakis, Stelios, Pimenidis, Elias and Petridis, Miltos (2019) A switching multi-level method for the long tail recommendation problem. Journal of Intelligent & Fuzzy Systems . pp. 1-10. ISSN 1064-1246 (Published online first) (doi:10.3233/jifs-179331)

[img]
Preview
PDF - Final accepted version (with author's formatting)
Download (659kB) | Preview

Abstract

Recommender systems are decision support systems that play an important part in generating a list of product or service recommendations for users based on the past experiences and interactions. The most popular recommendation method is Collaborative Filtering (CF) that is based on the users’ rating history to generate the recommendation. Although, recommender systems have been applied successfully in different areas such as e-Commerce and Social Networks, the popularity bias is still one of the challenges that needs to be further researched. Therefore, we propose a multi-level method that is based on a switching approach which solves the long tail recommendation problem (LTRP) when CF fails to find the target case. We have evaluated our method using two public datasets and the results show that it outperforms a number of bases lines and state-of-the-art alternatives with a further reduce of the recommendation error rates for items found in the long tail.

Item Type: Article
Additional Information: ISSN 1875-8967 (E)
Keywords (uncontrolled): General Engineering, Statistics and Probability, Artificial Intelligence
Research Areas: A. > School of Science and Technology > Computer and Communications Engineering
Item ID: 27242
Notes on copyright: The final publication is available at IOS Press through https://doi.org/10.3233/jifs-179331
Useful Links:
Depositing User: Jisc Publications Router
Date Deposited: 26 Jul 2019 08:14
Last Modified: 26 Jul 2019 09:06
URI: https://eprints.mdx.ac.uk/id/eprint/27242

Actions (login required)

Edit Item Edit Item

Full text downloads (NB count will be zero if no full text documents are attached to the record)

Downloads per month over the past year