A Regularization-Adaptive Non-negative Latent Factor Analysis-based Model for Recommender Systems

Jiufang Chen, Xin Luo, Meng Chu Zhou

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Non-negative latent factor analysis (NLFA) can high-efficiently extract useful information from high dimensional and sparse (HiDS) matrices often encountered in recommender systems (RSs). However, an NLFA-based model requires careful tuning of regularization coefficients, which is highly expensive in both time and computation. To address this issue, this study proposes an adaptive NLFA-based model whose regularization coefficients become self-Adaptive via particle swarm optimization. Experimental results on two HiDS matrices indicate that owing to such self-Adaptation, it outperforms an NLFA model in terms of both convergence rate and prediction accuracy for missing data estimation.

Original languageEnglish (US)
Title of host publicationProceedings of the 2020 IEEE International Conference on Human-Machine Systems, ICHMS 2020
EditorsGiancarlo Fortino, Fei-Yue Wang, Andreas Nurnberger, David Kaber, Rino Falcone, David Mendonca, Zhiwen Yu, Antonio Guerrieri
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781728158716
DOIs
StatePublished - Sep 2020
Event1st IEEE International Conference on Human-Machine Systems, ICHMS 2020 - Virtual, Rome, Italy
Duration: Sep 7 2020Sep 9 2020

Publication series

NameProceedings of the 2020 IEEE International Conference on Human-Machine Systems, ICHMS 2020

Conference

Conference1st IEEE International Conference on Human-Machine Systems, ICHMS 2020
CountryItaly
CityVirtual, Rome
Period9/7/209/9/20

All Science Journal Classification (ASJC) codes

  • Human-Computer Interaction

Keywords

  • High-Dimensional and Sparse Matrix
  • Non-negative Latent Factor Analysis
  • Particle Swarm Optimization
  • Recommender System
  • Self-Adaptive Model

Fingerprint Dive into the research topics of 'A Regularization-Adaptive Non-negative Latent Factor Analysis-based Model for Recommender Systems'. Together they form a unique fingerprint.

Cite this