基于图注意力网络的信息级联外源因素建模研究  

Modeling External Factors for Information Cascade Prediction via Graph Attention Network

作　　者：杨彩飘 鲍鹏[1] 李轩涯 YANG Caipiao;BAO Peng;LI Xuanya(School of Software Engineering,Beijing Jiaotong University,Beijing 100044,China;Baidu Online Network Technology Corporation,Beijing 100085,China)

机构地区：[1]北京交通大学软件学院,北京100044 [2]百度在线网络技术(北京)有限公司,北京100085

出　　处：《中文信息学报》2022年第5期163-172,共10页Journal of Chinese Information Processing

基　　金：国家自然科学基金(61702031);中国人工智能学会-华为MindSpore学术奖励基金;百度基金资助项目。

摘　　要：现有的信息级联预测方法忽略了外源因素对传播级联演化过程的影响以及个体在外源因素影响下的行为偏好,同时对底层的社交网络图结构信息的分析效果欠佳。为解决上述问题,该文提出基于图注意力网络的信息传播外源因素建模方法,利用图注意力机制提取社交图的结构信息,通过卷积神经网络对传播级联的时序信息进行分析,从而捕获外源因素的影响,利用循环神经网络对传播路径进行建模,最后在考虑到个体受外源因素的影响程度后进行级联预测。在Twitter、Douban和Memetracker三个真实数据集上的实验结果表明,相比于同类工作,该文提出的级联预测模型的性能较优。The current information cascade prediction methods ignore the evolution change of the diffusion cascade and the individual’s behavior preferences under the influence of external factors, as well as the graph structure of the social network. To address these issues, this paper proposes a method of modeling external factors in information diffusion based on graph attention network. The model applies graph attention mechanism to extract the underlying structure information in the social graphs. The convolutional neural networks are adopted to analyze the temporal information in the diffusion cascade and capture the external influence. A recurrent neural network is employed to model the diffusion path. Finally, the model utilizes different individual responses to the same external factors to predict the next node in the cascade. Experimental results on three real-world datasets from Twitter, Douban, and Memetracker show that the proposed model outperforms the state-of-the-art methods.

关 键 词：信息传播 级联预测 图注意力网络 循环神经网络 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]