Cargando…

A Hybrid Recommender System Based on Autoencoder and Latent Feature Analysis

A recommender system (RS) is highly efficient in extracting valuable information from a deluge of big data. The key issue of implementing an RS lies in uncovering users’ latent preferences on different items. Latent Feature Analysis (LFA) and deep neural networks (DNNs) are two of the most popular a...

Descripción completa

Detalles Bibliográficos
Autores principales: Guo, Shangzhi, Liao, Xiaofeng, Li, Gang, Xian, Kaiyi, Li, Yuhang, Liang, Cheng
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2023
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10378603/
https://www.ncbi.nlm.nih.gov/pubmed/37510009
http://dx.doi.org/10.3390/e25071062
_version_ 1785079808941621248
author Guo, Shangzhi
Liao, Xiaofeng
Li, Gang
Xian, Kaiyi
Li, Yuhang
Liang, Cheng
author_facet Guo, Shangzhi
Liao, Xiaofeng
Li, Gang
Xian, Kaiyi
Li, Yuhang
Liang, Cheng
author_sort Guo, Shangzhi
collection PubMed
description A recommender system (RS) is highly efficient in extracting valuable information from a deluge of big data. The key issue of implementing an RS lies in uncovering users’ latent preferences on different items. Latent Feature Analysis (LFA) and deep neural networks (DNNs) are two of the most popular and successful approaches to addressing this issue. However, both the LFA-based and the DNNs-based models have their own distinct advantages and disadvantages. Consequently, relying solely on either the LFA or DNN-based models cannot ensure optimal recommendation performance across diverse real-world application scenarios. To address this issue, this paper proposes a novel hybrid recommendation model that combines Autoencoder and LFA techniques, termed AutoLFA. The main idea of AutoLFA is two-fold: (1) It leverages an Autoencoder and an LFA model separately to construct two distinct recommendation models, each residing in a unique metric representation space with its own set of strengths; and (2) it integrates the Autoencoder and LFA model using a customized self-adaptive weighting strategy, thereby capitalizing on the merits of both approaches. To evaluate the proposed AutoLFA model, extensive experiments on five real recommendation datasets are conducted. The results demonstrate that AutoLFA achieves significantly better recommendation performance than the seven related state-of-the-art models.
format Online
Article
Text
id pubmed-10378603
institution National Center for Biotechnology Information
language English
publishDate 2023
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-103786032023-07-29 A Hybrid Recommender System Based on Autoencoder and Latent Feature Analysis Guo, Shangzhi Liao, Xiaofeng Li, Gang Xian, Kaiyi Li, Yuhang Liang, Cheng Entropy (Basel) Article A recommender system (RS) is highly efficient in extracting valuable information from a deluge of big data. The key issue of implementing an RS lies in uncovering users’ latent preferences on different items. Latent Feature Analysis (LFA) and deep neural networks (DNNs) are two of the most popular and successful approaches to addressing this issue. However, both the LFA-based and the DNNs-based models have their own distinct advantages and disadvantages. Consequently, relying solely on either the LFA or DNN-based models cannot ensure optimal recommendation performance across diverse real-world application scenarios. To address this issue, this paper proposes a novel hybrid recommendation model that combines Autoencoder and LFA techniques, termed AutoLFA. The main idea of AutoLFA is two-fold: (1) It leverages an Autoencoder and an LFA model separately to construct two distinct recommendation models, each residing in a unique metric representation space with its own set of strengths; and (2) it integrates the Autoencoder and LFA model using a customized self-adaptive weighting strategy, thereby capitalizing on the merits of both approaches. To evaluate the proposed AutoLFA model, extensive experiments on five real recommendation datasets are conducted. The results demonstrate that AutoLFA achieves significantly better recommendation performance than the seven related state-of-the-art models. MDPI 2023-07-14 /pmc/articles/PMC10378603/ /pubmed/37510009 http://dx.doi.org/10.3390/e25071062 Text en © 2023 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Guo, Shangzhi
Liao, Xiaofeng
Li, Gang
Xian, Kaiyi
Li, Yuhang
Liang, Cheng
A Hybrid Recommender System Based on Autoencoder and Latent Feature Analysis
title A Hybrid Recommender System Based on Autoencoder and Latent Feature Analysis
title_full A Hybrid Recommender System Based on Autoencoder and Latent Feature Analysis
title_fullStr A Hybrid Recommender System Based on Autoencoder and Latent Feature Analysis
title_full_unstemmed A Hybrid Recommender System Based on Autoencoder and Latent Feature Analysis
title_short A Hybrid Recommender System Based on Autoencoder and Latent Feature Analysis
title_sort hybrid recommender system based on autoencoder and latent feature analysis
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10378603/
https://www.ncbi.nlm.nih.gov/pubmed/37510009
http://dx.doi.org/10.3390/e25071062
work_keys_str_mv AT guoshangzhi ahybridrecommendersystembasedonautoencoderandlatentfeatureanalysis
AT liaoxiaofeng ahybridrecommendersystembasedonautoencoderandlatentfeatureanalysis
AT ligang ahybridrecommendersystembasedonautoencoderandlatentfeatureanalysis
AT xiankaiyi ahybridrecommendersystembasedonautoencoderandlatentfeatureanalysis
AT liyuhang ahybridrecommendersystembasedonautoencoderandlatentfeatureanalysis
AT liangcheng ahybridrecommendersystembasedonautoencoderandlatentfeatureanalysis
AT guoshangzhi hybridrecommendersystembasedonautoencoderandlatentfeatureanalysis
AT liaoxiaofeng hybridrecommendersystembasedonautoencoderandlatentfeatureanalysis
AT ligang hybridrecommendersystembasedonautoencoderandlatentfeatureanalysis
AT xiankaiyi hybridrecommendersystembasedonautoencoderandlatentfeatureanalysis
AT liyuhang hybridrecommendersystembasedonautoencoderandlatentfeatureanalysis
AT liangcheng hybridrecommendersystembasedonautoencoderandlatentfeatureanalysis