SVD-gorank: Recommender system algorithm using SVD and gower's ranking

Recommender systems using ranking-oriented collaborative filtering are currently widely used. One widely used approach is a memory-based model with ranking orientation. Recently, a ranking algorithm that combines user rating values from SVD (Singular Value Decomposition) and user similarity values h...

Full description

Bibliographic Details
Main Authors: Saifudin, Ilham, Widiyaningtyas, Triyanna, Zaeni, Ilham Ari Elbaith, Aminuddin, Afrig
Format: Article
Language:English
Published: Institute of Electrical and Electronics Engineers Inc. 2025
Subjects:
Online Access:http://umpir.ump.edu.my/id/eprint/43836/
http://umpir.ump.edu.my/id/eprint/43836/1/SVD-gorank_Recommender%20system%20algorithm%20using%20SVD%20and%20gower%27s%20ranking.pdf
_version_ 1848826971163721728
author Saifudin, Ilham
Widiyaningtyas, Triyanna
Zaeni, Ilham Ari Elbaith
Aminuddin, Afrig
author_facet Saifudin, Ilham
Widiyaningtyas, Triyanna
Zaeni, Ilham Ari Elbaith
Aminuddin, Afrig
author_sort Saifudin, Ilham
building UMP Institutional Repository
collection Online Access
description Recommender systems using ranking-oriented collaborative filtering are currently widely used. One widely used approach is a memory-based model with ranking orientation. Recently, a ranking algorithm that combines user rating values from SVD (Singular Value Decomposition) and user similarity values has been proposed. The problem is that this algorithm is limited to only the rating weights used. This results in an accuracy value that can still be improved. Therefore, this research proposes a new collaborative filtering-based algorithm that combines the matrix factorisation method using SVD and the ranking method by utilising Gower's Coefficient similarity weight as an aggregation component known as the SVD-GoRank method. Experimental results using the MovieLens-100K, MovieLens-1M, Book-Crossing, Ciao, Epinions, Flixster, and MovieLens-10M datasets can provide the best accuracy results at the Top-N level, especially in the NDCG, MRR, Precision, Hit Rate, and Recall metrics, which are indicators important in recommendation systems that focus on the relevance of recommendations at the top of the list. Apart from that, the SVD-GoRank algorithm can also have efficient running time.
first_indexed 2025-11-15T03:53:17Z
format Article
id ump-43836
institution Universiti Malaysia Pahang
institution_category Local University
language English
last_indexed 2025-11-15T03:53:17Z
publishDate 2025
publisher Institute of Electrical and Electronics Engineers Inc.
recordtype eprints
repository_type Digital Repository
spelling ump-438362025-02-17T08:04:21Z http://umpir.ump.edu.my/id/eprint/43836/ SVD-gorank: Recommender system algorithm using SVD and gower's ranking Saifudin, Ilham Widiyaningtyas, Triyanna Zaeni, Ilham Ari Elbaith Aminuddin, Afrig QA75 Electronic computers. Computer science QA76 Computer software T Technology (General) TA Engineering (General). Civil engineering (General) Recommender systems using ranking-oriented collaborative filtering are currently widely used. One widely used approach is a memory-based model with ranking orientation. Recently, a ranking algorithm that combines user rating values from SVD (Singular Value Decomposition) and user similarity values has been proposed. The problem is that this algorithm is limited to only the rating weights used. This results in an accuracy value that can still be improved. Therefore, this research proposes a new collaborative filtering-based algorithm that combines the matrix factorisation method using SVD and the ranking method by utilising Gower's Coefficient similarity weight as an aggregation component known as the SVD-GoRank method. Experimental results using the MovieLens-100K, MovieLens-1M, Book-Crossing, Ciao, Epinions, Flixster, and MovieLens-10M datasets can provide the best accuracy results at the Top-N level, especially in the NDCG, MRR, Precision, Hit Rate, and Recall metrics, which are indicators important in recommendation systems that focus on the relevance of recommendations at the top of the list. Apart from that, the SVD-GoRank algorithm can also have efficient running time. Institute of Electrical and Electronics Engineers Inc. 2025 Article PeerReviewed pdf en cc_by_4 http://umpir.ump.edu.my/id/eprint/43836/1/SVD-gorank_Recommender%20system%20algorithm%20using%20SVD%20and%20gower%27s%20ranking.pdf Saifudin, Ilham and Widiyaningtyas, Triyanna and Zaeni, Ilham Ari Elbaith and Aminuddin, Afrig (2025) SVD-gorank: Recommender system algorithm using SVD and gower's ranking. IEEE Access, 13. pp. 19796-19827. ISSN 2169-3536. (Published) https://doi.org/10.1109/ACCESS.2025.3533558 https://doi.org/10.1109/ACCESS.2025.3533558
spellingShingle QA75 Electronic computers. Computer science
QA76 Computer software
T Technology (General)
TA Engineering (General). Civil engineering (General)
Saifudin, Ilham
Widiyaningtyas, Triyanna
Zaeni, Ilham Ari Elbaith
Aminuddin, Afrig
SVD-gorank: Recommender system algorithm using SVD and gower's ranking
title SVD-gorank: Recommender system algorithm using SVD and gower's ranking
title_full SVD-gorank: Recommender system algorithm using SVD and gower's ranking
title_fullStr SVD-gorank: Recommender system algorithm using SVD and gower's ranking
title_full_unstemmed SVD-gorank: Recommender system algorithm using SVD and gower's ranking
title_short SVD-gorank: Recommender system algorithm using SVD and gower's ranking
title_sort svd-gorank: recommender system algorithm using svd and gower's ranking
topic QA75 Electronic computers. Computer science
QA76 Computer software
T Technology (General)
TA Engineering (General). Civil engineering (General)
url http://umpir.ump.edu.my/id/eprint/43836/
http://umpir.ump.edu.my/id/eprint/43836/
http://umpir.ump.edu.my/id/eprint/43836/
http://umpir.ump.edu.my/id/eprint/43836/1/SVD-gorank_Recommender%20system%20algorithm%20using%20SVD%20and%20gower%27s%20ranking.pdf