The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
Pada masa kini, sistem pengesyor (RS) terus menarik perhatian daripada ahli akademik, dan penapisan kolaboratif (CF) ialah teknik yang paling berjaya untuk membina RS. Untuk mengatasi had yang wujud, yang dirujuk sebagai keterbatasan data dalam CF, pelbagai penyelesaian dicadangkan untuk memasukkan maklumat sosial tambahan ke dalam proses pengesyoran, seperti rangkaian amanah. Walau bagaimanapun, kaedah sedia ada mengalami integrasi data berbilang sumber (iaitu, gabungan maklumat sosial dan penilaian), yang merupakan asas untuk pengiraan persamaan pilihan pengguna. Untuk tujuan ini, kami mencadangkan kaedah penapisan kolaboratif sosial berdasarkan metrik amanah baru. Pertama, kami menggunakan Graph Convolutional Networks (GCNs) untuk mempelajari perkaitan antara maklumat sosial dan rating pengguna sambil mempertimbangkan struktur rangkaian sosial yang mendasari. Kedua, kami mengukur nilai amanah langsung antara jiran dengan mewakili data berbilang sumber sebagai penilaian pengguna pada item popular, dan kemudian mengira nilai amanah tidak langsung berdasarkan penyebaran kepercayaan. Ketiga, kami menggunakan semua nilai amanah untuk mewujudkan penyelarasan sosial dalam pemfaktoran matriks penarafan item pengguna untuk mengelakkan pemasangan berlebihan. Percubaan pada set data sebenar menunjukkan bahawa pendekatan kami mengatasi kaedah tercanggih yang lain dalam penggunaan data berbilang sumber untuk mengurangkan keterbatasan data.
Haitao XIE
Beijing United Information Center of Science-Technology-Economy
Qingtao FAN
Beijing Institute of Science and Technology Information
Qian XIAO
Beijing Institute of Graphic Communication
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Salinan
Haitao XIE, Qingtao FAN, Qian XIAO, "A Social Collaborative Filtering Method to Alleviate Data Sparsity Based on Graph Convolutional Networks" in IEICE TRANSACTIONS on Information,
vol. E103-D, no. 12, pp. 2611-2619, December 2020, doi: 10.1587/transinf.2019EDP7286.
Abstract: Nowadays recommender systems (RS) keep drawing attention from academia, and collaborative filtering (CF) is the most successful technique for building RS. To overcome the inherent limitation, which is referred to as data sparsity in CF, various solutions are proposed to incorporate additional social information into recommendation processes, such as trust networks. However, existing methods suffer from multi-source data integration (i.e., fusion of social information and ratings), which is the basis for similarity calculation of user preferences. To this end, we propose a social collaborative filtering method based on novel trust metrics. Firstly, we use Graph Convolutional Networks (GCNs) to learn the associations between social information and user ratings while considering the underlying social network structures. Secondly, we measure the direct-trust values between neighbors by representing multi-source data as user ratings on popular items, and then calculate the indirect-trust values based on trust propagations. Thirdly, we employ all trust values to create a social regularization in user-item rating matrix factorization in order to avoid overfittings. The experiments on real datasets show that our approach outperforms the other state-of-the-art methods on usage of multi-source data to alleviate data sparsity.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2019EDP7286/_p
Salinan
@ARTICLE{e103-d_12_2611,
author={Haitao XIE, Qingtao FAN, Qian XIAO, },
journal={IEICE TRANSACTIONS on Information},
title={A Social Collaborative Filtering Method to Alleviate Data Sparsity Based on Graph Convolutional Networks},
year={2020},
volume={E103-D},
number={12},
pages={2611-2619},
abstract={Nowadays recommender systems (RS) keep drawing attention from academia, and collaborative filtering (CF) is the most successful technique for building RS. To overcome the inherent limitation, which is referred to as data sparsity in CF, various solutions are proposed to incorporate additional social information into recommendation processes, such as trust networks. However, existing methods suffer from multi-source data integration (i.e., fusion of social information and ratings), which is the basis for similarity calculation of user preferences. To this end, we propose a social collaborative filtering method based on novel trust metrics. Firstly, we use Graph Convolutional Networks (GCNs) to learn the associations between social information and user ratings while considering the underlying social network structures. Secondly, we measure the direct-trust values between neighbors by representing multi-source data as user ratings on popular items, and then calculate the indirect-trust values based on trust propagations. Thirdly, we employ all trust values to create a social regularization in user-item rating matrix factorization in order to avoid overfittings. The experiments on real datasets show that our approach outperforms the other state-of-the-art methods on usage of multi-source data to alleviate data sparsity.},
keywords={},
doi={10.1587/transinf.2019EDP7286},
ISSN={1745-1361},
month={December},}
Salinan
TY - JOUR
TI - A Social Collaborative Filtering Method to Alleviate Data Sparsity Based on Graph Convolutional Networks
T2 - IEICE TRANSACTIONS on Information
SP - 2611
EP - 2619
AU - Haitao XIE
AU - Qingtao FAN
AU - Qian XIAO
PY - 2020
DO - 10.1587/transinf.2019EDP7286
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E103-D
IS - 12
JA - IEICE TRANSACTIONS on Information
Y1 - December 2020
AB - Nowadays recommender systems (RS) keep drawing attention from academia, and collaborative filtering (CF) is the most successful technique for building RS. To overcome the inherent limitation, which is referred to as data sparsity in CF, various solutions are proposed to incorporate additional social information into recommendation processes, such as trust networks. However, existing methods suffer from multi-source data integration (i.e., fusion of social information and ratings), which is the basis for similarity calculation of user preferences. To this end, we propose a social collaborative filtering method based on novel trust metrics. Firstly, we use Graph Convolutional Networks (GCNs) to learn the associations between social information and user ratings while considering the underlying social network structures. Secondly, we measure the direct-trust values between neighbors by representing multi-source data as user ratings on popular items, and then calculate the indirect-trust values based on trust propagations. Thirdly, we employ all trust values to create a social regularization in user-item rating matrix factorization in order to avoid overfittings. The experiments on real datasets show that our approach outperforms the other state-of-the-art methods on usage of multi-source data to alleviate data sparsity.
ER -