Browse DORAS
Browse Theses
Search
Latest Additions
Creative Commons License
Except where otherwise noted, content on this site is licensed for use under a:

Social impact retrieval: measuring author influence on information retrieval

Lanagan, James (2009) Social impact retrieval: measuring author influence on information retrieval. PhD thesis, Dublin City University.

Full text available as:

[img]
Preview
PDF - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
12Mb

Abstract

The increased presence of technologies collectively referred to as Web 2.0 mean the entire process of new media production and dissemination has moved away from an authorcentric approach. Casual web users and browsers are increasingly able to play a more active role in the information creation process. This means that the traditional ways in which information sources may be validated and scored must adapt accordingly. In this thesis we propose a new way in which to look at a user's contributions to the network in which they are present, using these interactions to provide a measure of authority and centrality to the user. This measure is then used to attribute an query-independent interest score to each of the contributions the author makes, enabling us to provide other users with relevant information which has been of greatest interest to a community of like-minded users. This is done through the development of two algorithms; AuthorRank and MessageRank. We present two real-world user experiments which focussed around multimedia annotation and browsing systems that we built; these systems were novel in themselves, bringing together video and text browsing, as well as free-text annotation. Using these systems as examples of real-world applications for our approaches, we then look at a larger-scale experiment based on the author and citation networks of a ten year period of the ACM SIGIR conference on information retrieval between 1997-2007. We use the citation context of SIGIR publications as a proxy for annotations, constructing large social networks between authors. Against these networks we show the effectiveness of incorporating user generated content, or annotations, to improve information retrieval.

Item Type:Thesis (PhD)
Date of Award:November 2009
Refereed:No
Supervisor(s):Smeaton, Alan F.
Uncontrolled Keywords:social networking;
Subjects:Computer Science > Information storage and retrieval systems
Computer Science > Information retrieval
Computer Science > Digital video
DCU Faculties and Centres:DCU Faculties and Schools > Faculty of Engineering and Computing > School of Computing
Research Initiatives and Centres > CLARITY: The Centre for Sensor Web Technologies
Use License:This item is licensed under a Creative Commons Attribution-NonCommercial-No Derivative Works 3.0 License. View License
Funders:Science Foundation Ireland
ID Code:14852
Deposited On:12 Nov 2009 10:20 by Alan F. Smeaton. Last Modified 12 Nov 2009 10:20

Download statistics

Archive Staff Only: edit this record