基于情境的社交网络信息传播链路预测研究  

Research on the Link Prediction of Social Network Information Dissemination Based on the Context

作　　者：郑建国[1] 朱君璇[1,2] 曹如中[2] 

机构地区：[1]东华大学旭日工商管理学院,上海200051 [2]上海工程技术大学管理学院,上海201620

出　　处：《情报理论与实践》2018年第6期94-99,共6页Information Studies:Theory & Application

基　　金：国家社会科学基金项目"非常规突发事件中社会化媒体不实信息的群体干预模式研究"的成果;项目编号:14BTQ026

摘　　要：[目的/意义]针对目前链路预测模型准确度不高的问题,提出将情境特征融入到传统的链路预测研究中,以此来提高链路预测的精确度。[过程/方法]在对情境特征分析的基础上结合朴素贝叶斯理论,构建了基于情境特征的兴趣加权贝叶斯链路预测模型,并应用腾讯微博数据对该模型的准确性进行验证。[结果/结论]由实验结果分析可知,融入情景特征以及考虑兴趣加权的链路预测模型相比于传统的链路预测模型精确度更高,更贴近现实的社交网络。[Purpose/significance] Aiming at the problem that the current link prediction model is not accurate enough,this paper proposes to integrate context feature into the traditional link prediction research so as to improve the accuracy of link prediction. [Method/process] Based on the analysis of contextual features and the naive Bayesian theory,the paper constructs the interest-weighted Bayesian link prediction model based on contextual features,and applies the model to Tencent Weibo data to verify its accuracy. [Result/conclusion] From the experimental results,it can be seen that the link prediction model that integrates the contextual features and the weighted interest is more accurate and closer to the actual social network than the traditional link prediction model.

关 键 词：情境 社交网络 信息传播 链路预测 

分 类 号：G206[文化科学—传播学]