CORC  > 北京大学  > 信息科学技术学院
Using Structural Features to Characterize Social Ties
Zuo, Yang ; Zhang, Kan
2016
关键词social networks social ties structural features logistic regression model NETWORKS STRENGTH
英文摘要Social ties are crucial to understand group behaviors in social networks. Since users rarely label their friends explicitly in social networks, characterizing social ties and understanding their strength of these ties is a critical problem. In this paper, we apply logistic regression model to infer the strength of social ties based on the structural features of social networks, and we produce two new features: cliqueness and linkness. In addition, we use the Strong Triadic Closure principle as a global constraint. We test our model on four different datasets: Facebook, American College Football, Les Miserables and Zachary's Karate Club. Experimental results demonstrate that the performance of our model with structural features is satisfactory.; State Key Development Program of Basic Research of China (973) [2013cb329600]; National Natural Science Foundation of China [61372191, 61572492]; CPCI-S(ISTP); 235-242
语种英语
出处1st IEEE International Conference on Data Science in Cyberspace (DSC)
DOI标识10.1109/DSC.2016.63
内容类型其他
源URL[http://ir.pku.edu.cn/handle/20.500.11897/470083]  
专题信息科学技术学院
推荐引用方式
GB/T 7714
Zuo, Yang,Zhang, Kan. Using Structural Features to Characterize Social Ties. 2016-01-01.
个性服务
查看访问统计
相关权益政策
暂无数据
收藏/分享
所有评论 (0)
暂无评论
 

除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。


©版权所有 ©2017 CSpace - Powered by CSpace