Balancing Information Exposure in Social Networks

Summary:

In this work, we address the problem of balancing the information exposure in a social network. We assume that two opposing campaigns (or viewpoints) are present in the network, and network nodes have different preferences towards these campaigns. Our goal is to find two sets of nodes to employ in the respective campaigns, so that the overall information-exposure balance in the network is maximized. The camera ready version, along with the supplementary materials can be found here. A poster of our paper can be found here.

Datasets:

We collected 6 datasets from Twitter.

    Dataset description and keywords used to identify the sides. The datasets are available on Github.
    Dataset Side1 Side2
    USElections Pro-Hillary Pro-Trump
    Description: Tweets containing hashtags and keywords identifying the USElections, such as
    #uselections, #trump2016, #hillary2016, etc. Collected using Twitter 1% sample for 2 weeks in September 2016    		 RT @hillaryclinton, #hillary2016, #clintonkaine2016, #imwithher RT @realdonaldtrump, #makeamericagreatagain, #trumppence16, #trump2016
    Brexit Pro-Remain Pro-Leave
    Description: Tweets containing hashtags #brexit, #voteremain, #voteleave, #eureferendum for all of June 2016, from the 1% Twitter sample. #voteremain, #strongerin, #remain, #remaineu, #votein #voteleave, #strongerout, #leaveeu, #takecontrol, #leave, #voteout
    Abortion Pro-Abortion Anti-Abortion
    Description: Tweets containing hashtags #abortion, #prolife, #prochoice, #anti-abortion, #pro-abortion, #plannedparenthood from Oct 2011 to Aug 2016. RT @thinkprogress, RT @komenforthecure, RT @mentalabortions, #waronwomen, #nbprochoice, #prochoice, #standwithpp, #reprorights RT @stevenertelt, RT @lifenewshq, #praytoendabortion, #prolifeyouth, #prolife, #defundplannedparenthood, #defundpp, #unbornlivesmatter
    Obamacare
    		 Pro-Obamacare Anti-Obamacare
    Description: Tweets containing hashtags #obamacare, and #aca from Oct 2011 to Aug 2016. RT @barackobama, RT @lolgop, RT @charlespgarcia, RT @defendobamacare, RT @thinkprogress, #obamacares, #enoughalready, #uniteblue RT @sentedcruz, RT @realdonaldtrump, RT @mittromney, RT @breitbartnews, RT @tedcruz, #defundobamacare, #makedclisten, #fullrepeal, #dontfundit
    Fracking
    		 Pro-Fracking
    		 Anti-Fracking
    Description: Tweets containing hashtags and keywords #fracking, 'hydraulic fracturing', 'shale', 'horizontal drilling', from Oct 2011 to Aug 2016. RT @shalemarkets, RT @energyindepth, RT @shalefacts, #fracknation, #frackingez, #oilandgas, #greatgasgala, #shalegas RT @greenpeaceuk, RT @greenpeace, RT @ecowatch, #environment, #banfracking, #keepitintheground, #dontfrack, #globalfrackdown, #stopthefrackattack
    iPhone vs. Samsung
    		 Pro-iPhone Pro-Samsung
    Description: Tweets containing hashtags #iphone, and #samsung from April (release of Samsung Galaxy S7), and September 2015 (release of iPhone 7). #iphone #samsung

Code:

The code used in the paper is on Github.

Word Clouds:

To evaluate the results qualitatively, we developed word clouds from the profiles of the seeds initially considered, and the seeds we obtain using our algorithm. The results can be found here.

Contact: