MGNN: mutualistic graph neural network for joint friend and item recommendation

Yang Xiao; Lina Yao; Qingqi Pei; Xianzhi Wang; Jian Yang; Quan Z. Sheng

doi:10.1109/MIS.2020.2988925

MGNN: mutualistic graph neural network for joint friend and item recommendation

Yang Xiao, Lina Yao, Qingqi Pei, Xianzhi Wang, Jian Yang, Quan Z. Sheng

School of Computing

Research output: Contribution to journal › Article › peer-review

16 Citations (Scopus)

44 Downloads (Pure)

Abstract

Many social studies and practical cases suggest that people's consumption behaviors and social behaviors are not isolated but interrelated in social network services. However, most existing research either predicts users' consumption preferences or recommends friends to users without dealing with them simultaneously. We propose a holistic approach to predict users' preferences on friends and items jointly and thereby make better recommendations. To this end, we design a graph neural network that incorporates a mutualistic mechanism to model the mutual reinforcement relationship between users' consumption behaviors and social behaviors. Our experiments on the two-real world datasets demonstrate the effectiveness of our approach in both social recommendation and link prediction.

Original language	English
Pages (from-to)	7-17
Number of pages	11
Journal	IEEE Intelligent Systems
Volume	35
Issue number	5
Early online date	24 Apr 2020
DOIs	https://doi.org/10.1109/MIS.2020.2988925
Publication status	Published - Sept 2020

Bibliographical note

Version archived for private and non-commercial use with the permission of the author/s and according to publisher conditions. For further rights please contact the publisher.

Keywords

Graph Neural Networks
Joint Recommendation
Mutualistic Model
Social Networks

Access to Document

10.1109/MIS.2020.2988925

Accepted author manucsriptAccepted author manuscript, 1.41 MB

Cite this

@article{307152b0d26649d2b46e91b8dd36c1e1,

title = "MGNN: mutualistic graph neural network for joint friend and item recommendation",

abstract = "Many social studies and practical cases suggest that people's consumption behaviors and social behaviors are not isolated but interrelated in social network services. However, most existing research either predicts users' consumption preferences or recommends friends to users without dealing with them simultaneously. We propose a holistic approach to predict users' preferences on friends and items jointly and thereby make better recommendations. To this end, we design a graph neural network that incorporates a mutualistic mechanism to model the mutual reinforcement relationship between users' consumption behaviors and social behaviors. Our experiments on the two-real world datasets demonstrate the effectiveness of our approach in both social recommendation and link prediction.",

keywords = "Graph Neural Networks, Joint Recommendation, Mutualistic Model, Social Networks",

author = "Yang Xiao and Lina Yao and Qingqi Pei and Xianzhi Wang and Jian Yang and Sheng, {Quan Z.}",

note = "Version archived for private and non-commercial use with the permission of the author/s and according to publisher conditions. For further rights please contact the publisher.",

year = "2020",

month = sep,

doi = "10.1109/MIS.2020.2988925",

language = "English",

volume = "35",

pages = "7--17",

journal = "IEEE Intelligent Systems",

issn = "1541-1672",

publisher = "Institute of Electrical and Electronics Engineers (IEEE)",

number = "5",

}

TY - JOUR

T1 - MGNN

T2 - mutualistic graph neural network for joint friend and item recommendation

AU - Xiao, Yang

AU - Yao, Lina

AU - Pei, Qingqi

AU - Wang, Xianzhi

AU - Yang, Jian

AU - Sheng, Quan Z.

N1 - Version archived for private and non-commercial use with the permission of the author/s and according to publisher conditions. For further rights please contact the publisher.

PY - 2020/9

Y1 - 2020/9

N2 - Many social studies and practical cases suggest that people's consumption behaviors and social behaviors are not isolated but interrelated in social network services. However, most existing research either predicts users' consumption preferences or recommends friends to users without dealing with them simultaneously. We propose a holistic approach to predict users' preferences on friends and items jointly and thereby make better recommendations. To this end, we design a graph neural network that incorporates a mutualistic mechanism to model the mutual reinforcement relationship between users' consumption behaviors and social behaviors. Our experiments on the two-real world datasets demonstrate the effectiveness of our approach in both social recommendation and link prediction.

AB - Many social studies and practical cases suggest that people's consumption behaviors and social behaviors are not isolated but interrelated in social network services. However, most existing research either predicts users' consumption preferences or recommends friends to users without dealing with them simultaneously. We propose a holistic approach to predict users' preferences on friends and items jointly and thereby make better recommendations. To this end, we design a graph neural network that incorporates a mutualistic mechanism to model the mutual reinforcement relationship between users' consumption behaviors and social behaviors. Our experiments on the two-real world datasets demonstrate the effectiveness of our approach in both social recommendation and link prediction.

KW - Graph Neural Networks

KW - Joint Recommendation

KW - Mutualistic Model

KW - Social Networks

UR - http://www.scopus.com/inward/record.url?scp=85084079380&partnerID=8YFLogxK

U2 - 10.1109/MIS.2020.2988925

DO - 10.1109/MIS.2020.2988925

M3 - Article

AN - SCOPUS:85084079380

SN - 1541-1672

VL - 35

SP - 7

EP - 17

JO - IEEE Intelligent Systems

JF - IEEE Intelligent Systems

IS - 5

ER -

MGNN: mutualistic graph neural network for joint friend and item recommendation

Abstract

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this