Design of Confidence-Integrated Denoising Auto-Encoder for Personalized Top-N Recommender Systems

Zeshan Aslam Khan, Naveed Ishtiaq Chaudhary (), Waqar Ali Abbasi, Sai Ho Ling and Muhammad Asif Zahoor Raja
Additional contact information
Zeshan Aslam Khan: Department of Electrical and Computer Engineering, International Islamic University, Islamabad 44000, Pakistan
Naveed Ishtiaq Chaudhary: Future Technology Research Center, National Yunlin University of Science and Technology, 123 University Road, Section 3, Douliou, Yunlin 64002, Taiwan
Waqar Ali Abbasi: Department of Electrical and Computer Engineering, International Islamic University, Islamabad 44000, Pakistan
Sai Ho Ling: Faculty of Engineering and IT, University of Technology Sydney, Ultimo 2007, Australia
Muhammad Asif Zahoor Raja: Future Technology Research Center, National Yunlin University of Science and Technology, 123 University Road, Section 3, Douliou, Yunlin 64002, Taiwan

Mathematics, 2023, vol. 11, issue 3, 1-25

Abstract: A recommender system not only “gains users’ confidence” but also helps them in other ways, such as reducing their time spent and effort. To gain users’ confidence, one of the main goals of recommender systems in an e-commerce industry is to estimate the users’ interest by tracking the users’ transactional behavior to provide a fast and highly related set of top recommendations out of thousands of products. The standard ranking-based models, i.e., the denoising auto-encoder (DAE) and collaborative denoising auto-encoder (CDAE), exploit positive-only feedback without utilizing the ratings’ ranks for the full set of observed ratings. To confirm the rank of observed ratings (either low or high), a confidence value for each rating is required. Hence, an improved, confidence-integrated DAE is proposed to enhance the performance of the standard DAE for solving recommender systems problems. The correctness of the proposed method is authenticated using two standard MovieLens datasets such as ML-1M and ML-100K. The proposed study acts as a vital contribution for the design of an efficient, robust, and accurate algorithm by learning prominent latent features used for fast and accurate recommendations. The proposed model outperforms the state-of-the-art methods by achieving improved P@10, R@10, NDCG@10, and MAP scores.

Keywords: e-commerce; recommender systems; collaborative; confidence; denoising; auto-encoder (search for similar items in EconPapers)
JEL-codes: C (search for similar items in EconPapers)
Date: 2023
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://www.mdpi.com/2227-7390/11/3/761/pdf (application/pdf)
https://www.mdpi.com/2227-7390/11/3/761/ (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:gam:jmathe:v:11:y:2023:i:3:p:761-:d:1055549

Access Statistics for this article

Mathematics is currently edited by Ms. Emma He

More articles in Mathematics from MDPI
Bibliographic data for series maintained by MDPI Indexing Manager ().