Files in this item

FilesDescriptionFormat

application/pdf

application/pdf2pt14_Xia-Thesis.pdf (404kB)
(no description provided)PDF

Description

Title:Internal/External information access and information diffusion in social media
Author(s):Xia, Tian; Yu, Xing; Gao, Zheng; Gu, Yijun; Liu, Xiaozhong
Subject(s):Social media
Information adoption
Information diffusion
Information access
Weibo
Abstract:As social media platform not only provide infrastructure but also actively perform algorithmic curation for profit and user experience, it leads to an information filter bubble phenomenon: users are trapped in their own personalized bubble and are exposed only to the opinions that conform their beliefs and interests, thus potentially creating social polarization and information islands. However, filter bubbles hardly restrict all the users in a large social network, some information explorers can break the bubble and bring external global knowledge back to the internal network. In this paper, we investigate this assumption via hashtag adoption prediction. First, we construct a heterogeneous graph and extract 17 features to describe the event of hashtag adoption. Then, we generate learning instances and train a lasso regression model to do prediction. Preliminary results show that information explorers are more likely to adopt new hashtags than others, thereby more internal and external information can be diffused via these special users.
Issue Date:2017
Publisher:iSchools
Citation Info:Xia, T., Yu, X., Gao, Z., Gu, Y. & Liu, X. (2017). Internal/External Information Access and Information Diffusion in Social Media. In iConference Proceedings 2017, Vol. 2 (pp. 129-133). https://doi.org/10.9776/17235
Series/Report:iConference 2017 Proceedings Vol. 2
Genre:Conference Paper / Presentation
Type:Text
Language:English
URI:http://hdl.handle.net/2142/98867
DOI:https://doi.org/10.9776/17235
Rights Information:Copyright 2017 is held by the authors.
Date Available in IDEALS:2017-12-05


This item appears in the following Collection(s)

Item Statistics