Utilizing Social Networks for User Model Priming: User Attitudes


Published on

Published in: Education
  • Be the first to comment

  • Be the first to like this

No Downloads
Total views
On SlideShare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide

Utilizing Social Networks for User Model Priming: User Attitudes

  1. 1. Utilizing Social Networks for User Model Priming:User Attitudes5. Department of Psychiatry, School ofMedicine, Trinity College, Dublin, IrelandDONOGHUG@tcd.ie4. EmpowerTheUser, Trinity Technology &Enterprise Campus, The Tower, Dublin, Irelanddeclan.dagger@empowertheuser.com3. Delft University ofTechnology, the Netherlandsc.hauff@tudelft.nl2. Knowledge Technologies Institute,Graz University of Technology, Austria{gudrun.wesiak,christina.steiner}@tugraz.at1. KDEG, School of Computer Science andStatistics, Trinity College, Dublin, Ireland{mooread,owen.conlan}@scss.tcd.ieAdam Moore1, Gudrun Wesiak2, Christina M. Steiner2, Claudia Hauff3, Declan Dagger4, Gary Donohoe5 & Owen Conlan1The research leading to these results has receivedfunding from the European Communitys SeventhFramework Program (FP7/2007-2013) under grantagreement no 257831 (ImREAL project).Research has shown that some user characteristics can be accuratelyinferred from users’ digital traces. This study addresses a crucial questionfor practical applications of this approach:Are users actually willing to provide their social networkprofiles and how do they perceive this?An empirical study conducted with medical students shows that althoughparticipants are using social networks, they are reluctant about providingtheir identities and consider these portals rather private.Abstractisthisok?We asked 152 third year medical students abouttheir social network usage via onlinequestionnaire – 95 answered.What social networks do they useand who do they expect their poststo be read by?81% of the students use Facebook, Twitter isused by only 20%, LinkedIn by 5.3% and Flickr byonly 1 person.The expected audience (by percentage) for eachsocial network’s posts is shown alongside.Social NetworksWe asked the students to provide us with their social network ID, forexample, their Twitter handle.When asked to provide their user identity for any socialnetwork, most declined.From the 77 Facebook users only 11 (14.3%) provided theirusername, Twitter and LinkedIn usernames were provided by 3 (15.8%)and one person (20%), respectively. In total IDs from 13 persons wereprovided.When asked to explain, 60 participants responded, of which 49 or 83%explained why they did not want to provide their SN-ID, whereas theremaining persons gave a reason for providing their ID. All answers wereanalyzed and aggregated to the 6 categories shown alongside.PrivacyResearch has shown that some user characteristics can be accurately inferred from users’ digital traces. However, when directlyquestioned about this approach, our users are reluctant to disclose their identities within these networks and express discomfort.In the follow up simulation, when offered the opportunity to enable enhancement to the learningexperience by adding their Twitter handle to their profile, not one student provided it.Future research needs to focus on conclusive ways to convey the benefits for the users of these approaches and to give them morecontrol and insight on the actually utilized body of information.ConclusionsIntendedAudienceProviding SNID