A Hybrid Model Based on LFM and BiGRU Toward Research Paper Recommendation

To improve the accuracy of user implicit rating prediction, we combine the traditional latent factor model (LFM) and bidirectional gated recurrent unit neural network (BiGRU) model to propose a hybrid model that deeply mines the latent semantics in the unstructured content of the text and generates...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Xu Zhao, Hui Kang, Tie Feng, Chenkun Meng, Ziqing Nie
Formato:	article
Lenguaje:	EN
Publicado:	IEEE 2020
Materias:	Recommender systems deep learning LFM BiGRU user attention Electrical engineering. Electronics. Nuclear engineering TK1-9971
Acceso en línea:	https://doaj.org/article/41802f3dbc0e436ebed02f51ce2da93e
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:41802f3dbc0e436ebed02f51ce2da93e
record_format	dspace
spelling	oai:doaj.org-article:41802f3dbc0e436ebed02f51ce2da93e2021-11-19T00:05:57ZA Hybrid Model Based on LFM and BiGRU Toward Research Paper Recommendation2169-353610.1109/ACCESS.2020.3031281https://doaj.org/article/41802f3dbc0e436ebed02f51ce2da93e2020-01-01T00:00:00Zhttps://ieeexplore.ieee.org/document/9224616/https://doaj.org/toc/2169-3536To improve the accuracy of user implicit rating prediction, we combine the traditional latent factor model (LFM) and bidirectional gated recurrent unit neural network (BiGRU) model to propose a hybrid model that deeply mines the latent semantics in the unstructured content of the text and generates a more accurate rating matrix. First, we utilize the user’s historical behavior (favorites records) to build a user rating matrix and decompose the matrix to obtain the latent factor vectors of users and literature. We also apply the BERT model for word embedding of the research papers to obtain the sequence of word vectors. Then, we apply the BiGRU with the user attention mechanism to mine the research paper textual content and to generate the new literature latent feature vectors that are used to replace the original literature latent factor vectors decomposed from the rating matrix. Finally, a new rating matrix is generated to obtain users’ ratings of noninteractive research papers and to generate the recommendation list according to the user latent factor vector. We design experiments on the real datasets and verify that the research paper recommendation model is superior to traditional recommendation models in terms of precision, recall, F1-value, coverage, popularity and diversity.Xu ZhaoHui KangTie FengChenkun MengZiqing NieIEEEarticleRecommender systemsdeep learningLFMBiGRUuser attentionElectrical engineering. Electronics. Nuclear engineeringTK1-9971ENIEEE Access, Vol 8, Pp 188628-188640 (2020)
institution	DOAJ
collection	DOAJ
language	EN
topic	Recommender systems deep learning LFM BiGRU user attention Electrical engineering. Electronics. Nuclear engineering TK1-9971
spellingShingle	Recommender systems deep learning LFM BiGRU user attention Electrical engineering. Electronics. Nuclear engineering TK1-9971 Xu Zhao Hui Kang Tie Feng Chenkun Meng Ziqing Nie A Hybrid Model Based on LFM and BiGRU Toward Research Paper Recommendation
description	To improve the accuracy of user implicit rating prediction, we combine the traditional latent factor model (LFM) and bidirectional gated recurrent unit neural network (BiGRU) model to propose a hybrid model that deeply mines the latent semantics in the unstructured content of the text and generates a more accurate rating matrix. First, we utilize the user’s historical behavior (favorites records) to build a user rating matrix and decompose the matrix to obtain the latent factor vectors of users and literature. We also apply the BERT model for word embedding of the research papers to obtain the sequence of word vectors. Then, we apply the BiGRU with the user attention mechanism to mine the research paper textual content and to generate the new literature latent feature vectors that are used to replace the original literature latent factor vectors decomposed from the rating matrix. Finally, a new rating matrix is generated to obtain users’ ratings of noninteractive research papers and to generate the recommendation list according to the user latent factor vector. We design experiments on the real datasets and verify that the research paper recommendation model is superior to traditional recommendation models in terms of precision, recall, F1-value, coverage, popularity and diversity.
format	article
author	Xu Zhao Hui Kang Tie Feng Chenkun Meng Ziqing Nie
author_facet	Xu Zhao Hui Kang Tie Feng Chenkun Meng Ziqing Nie
author_sort	Xu Zhao
title	A Hybrid Model Based on LFM and BiGRU Toward Research Paper Recommendation
title_short	A Hybrid Model Based on LFM and BiGRU Toward Research Paper Recommendation
title_full	A Hybrid Model Based on LFM and BiGRU Toward Research Paper Recommendation
title_fullStr	A Hybrid Model Based on LFM and BiGRU Toward Research Paper Recommendation
title_full_unstemmed	A Hybrid Model Based on LFM and BiGRU Toward Research Paper Recommendation
title_sort	hybrid model based on lfm and bigru toward research paper recommendation
publisher	IEEE
publishDate	2020
url	https://doaj.org/article/41802f3dbc0e436ebed02f51ce2da93e
work_keys_str_mv	AT xuzhao ahybridmodelbasedonlfmandbigrutowardresearchpaperrecommendation AT huikang ahybridmodelbasedonlfmandbigrutowardresearchpaperrecommendation AT tiefeng ahybridmodelbasedonlfmandbigrutowardresearchpaperrecommendation AT chenkunmeng ahybridmodelbasedonlfmandbigrutowardresearchpaperrecommendation AT ziqingnie ahybridmodelbasedonlfmandbigrutowardresearchpaperrecommendation AT xuzhao hybridmodelbasedonlfmandbigrutowardresearchpaperrecommendation AT huikang hybridmodelbasedonlfmandbigrutowardresearchpaperrecommendation AT tiefeng hybridmodelbasedonlfmandbigrutowardresearchpaperrecommendation AT chenkunmeng hybridmodelbasedonlfmandbigrutowardresearchpaperrecommendation AT ziqingnie hybridmodelbasedonlfmandbigrutowardresearchpaperrecommendation
_version_	1718420664883347456

A Hybrid Model Based on LFM and BiGRU Toward Research Paper Recommendation

Ejemplares similares