基于多质心异质图学习的社交网络用户建模(User Representation Learning based on Multi-centroid Heterogeneous Graph Neural Networks)

Shangyi Ning (宁上毅), Guanying Li (李冠颖), Qin Chen (陈琴), Zengfeng Huang (黄增峰), Baohua Zhou (周葆华), Zhongyu Wei (魏忠钰)


Abstract
用户建模已经引起了学术界和工业界的广泛关注。现有的方法大多侧重于将用户间的人际关系融入社区,而用户生成的内容(如帖子)却没有得到很好的研究。在本文中,我们通过实际舆情传播相关的分析表明,在舆情传播过程中对用户属性进行研究的重要作用,并且提出了用户资料数据的筛选方法。同时,我们提出了一种通过异构多质心图池为用户捕获更多不同社区特征的建模。我们首先构造了一个由用户和关键字组成的异质图,并在其上采用了一个异质图神经网络。为了方便用户建模的图表示,提出了一种多质心图池化机制,将多质心的集群特征融入到表示学习中。在三个基准数据集上的大量实验表明了该方法的有效性。
Anthology ID:
2021.ccl-1.74
Volume:
Proceedings of the 20th Chinese National Conference on Computational Linguistics
Month:
August
Year:
2021
Address:
Huhhot, China
Venue:
CCL
SIG:
Publisher:
Chinese Information Processing Society of China
Note:
Pages:
825–836
Language:
Chinese
URL:
https://aclanthology.org/2021.ccl-1.74
DOI:
Bibkey:
Cite (ACL):
Shangyi Ning, Guanying Li, Qin Chen, Zengfeng Huang, Baohua Zhou, and Zhongyu Wei. 2021. 基于多质心异质图学习的社交网络用户建模(User Representation Learning based on Multi-centroid Heterogeneous Graph Neural Networks). In Proceedings of the 20th Chinese National Conference on Computational Linguistics, pages 825–836, Huhhot, China. Chinese Information Processing Society of China.
Cite (Informal):
基于多质心异质图学习的社交网络用户建模(User Representation Learning based on Multi-centroid Heterogeneous Graph Neural Networks) (Ning et al., CCL 2021)
Copy Citation:
PDF:
https://preview.aclanthology.org/ingestion-script-update/2021.ccl-1.74.pdf