Link to the University of Pittsburgh Homepage
Link to the University Library System Homepage Link to the Contact Us Form

A Study of Information Fragment Association in Information Management and Retrieval Applications

Ioannides, Demetrios (2007) A Study of Information Fragment Association in Information Management and Retrieval Applications. Doctoral Dissertation, University of Pittsburgh. (Unpublished)

Primary Text

Download (6MB) | Preview


As we strive to identify useful information sifting through the vast number of resources available to us, we often find that the desired information is residing in a small section within a larger body of content which does not necessarily contain similar information. This can make this Information Fragment difficult to find. A Web search engine may not provide a good ranking to a page of unrelated content if it contains only a very small yet invaluable piece of relevant information. This means that our processes often fail to bring together related Information Fragments. We can easily conceive of two Information Fragments which according to a scholar bear a strong association with each other, yet contain no common keywords enabling them to be collocated by a keyword search.This dissertation attempts to address this issue by determining the benefits of enhancing information management and retrieval applications by providing users with the capability of establishing and storing associations between Information Fragments. It estimates the extent to which the efficiency and quality of information retrieval can be improved if users are allowed to capture mental associations they form while reading Information Fragments and share these associations with others using a functional registry-based design. In order to test these benefits three subject groups were recruited and assigned tasks involving Information Fragments. The first two tasks compared the performance and usability of a mainstream social bookmarking tool with a tool enhanced with Information Fragment Association capabilities. The tests demonstrated that the use of Information Fragment Association offers significant advantages both in the efficiency of retrieval and user satisfaction. Analysis of the results of the third task demonstrated that a mainstream Web search engine performed poorly in collocating interrelated fragments when a query designed to retrieve the one of these fragments was submitted. The fourth task demonstrated that Information Fragment Association improves the precision and recall of searches performed on Information Fragment datasets.The results of this study indicate that mainstream information management and retrieval applications provide inadequate support for Information Fragment retrieval and that their enhancement with Information Fragment Association capabilities would be beneficial.


Social Networking:
Share |


Item Type: University of Pittsburgh ETD
Status: Unpublished
CreatorsEmailPitt UsernameORCID
Ioannides, Demetriosdei1@pitt.eduDEI1
ETD Committee:
TitleMemberEmail AddressPitt UsernameORCID
Committee ChairDaqing, Hedaqing@sis.pitt.eduDAH44
Committee MemberDavid,
Committee MemberMichael, Springspring@pitt.eduSPRING
Committee MemberChristinger, Tomerctomer@sis.pitt.eduCTOMER
Date: 20 September 2007
Date Type: Completion
Defense Date: 3 August 2007
Approval Date: 20 September 2007
Submission Date: 16 August 2007
Access Restriction: No restriction; Release the ETD for access worldwide immediately.
Institution: University of Pittsburgh
Schools and Programs: School of Information Sciences > Library and Information Science
Degree: PhD - Doctor of Philosophy
Thesis Type: Doctoral Dissertation
Refereed: Yes
Uncontrolled Keywords: Hypertext; Information Entity; Information Management Tools; Bookmarking Tools; Commentary Systems; Information Fragment; Information Fragment Association; Information Retrieval Tools; Open Hypermedia Systems; Information System Frameworks; Information Architecture
Other ID:, etd-08162007-220245
Date Deposited: 10 Nov 2011 19:59
Last Modified: 15 Nov 2016 13:49


Monthly Views for the past 3 years

Plum Analytics

Actions (login required)

View Item View Item