Show simple item record

dc.contributor.authorChen, Liangzheen_US
dc.contributor.authorPrakash, B. Adityaen_US
dc.date.accessioned2018-04-09T13:13:35Z
dc.date.available2018-04-09T13:13:35Z
dc.date.issued2018-04-09
dc.identifier.urihttp://hdl.handle.net/10919/82748
dc.description.abstractMicroblogging websites, like Twitter and Weibo, are used by billions of people to create and spread information. This activity depends on various factors such as the friendship links between users, their topic interests and social influence between them. Making sense of these behaviors is very important for fully understanding and utilizing these platforms. Most prior work on modeling social-media either ignores the effect of social influence, or considers its effect only on link formation or post generation. In contrast, in this paper we propose POLIM, which jointly models the effect of influence on both link and post generation, leveraging weak supervision. We also give POLIM-FIT, an efficient parallel inference algorithm for POLIM which scales to large datasets. In our experiments on a large tweets corpus, we detect meaningful topical communities, celebrities, as well as the influence strengths patterns among them. Further, we find that there are significant portions of posts and links that are caused by influence, and this portion increases when the data focuses on a specific event. We also show that differentiating and identifying these influenced content benefits other quantitative downstream tasks as well, like predicting future tweets and link formation.en_US
dc.language.isoen_USen_US
dc.publisherDepartment of Computer Science, Virginia Polytechnic Institute & State Universityen_US
dc.relation.ispartofComputer Science Technical Reportsen_US
dc.subjectData and Text Miningen_US
dc.titleModeling Influence using Weak Supervision: A joint Link and Post-level Analysisen_US
dc.typeTechnical reporten_US
dc.identifier.trnumberTR-18-03en_US
dc.type.dcmitypeTexten_US


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record