|
Type of Document Dissertation Author Yi, Myongho Author's Email Address myonghoyi@gmail.com URN etd-04102006-100631 Title User Performance Using an Ontology-Driven Information Retrieval (ONTOIR)System Degree Doctor of Philosophy Department Information Studies, College of Advisory Committee
Advisor Name Title Kathleen Burnett Committee Chair Akihito Kamata Committee Member Darrell Burke Committee Member Gary Burnett Committee Member Kyunghye Kim Committee Member Keywords
- Information Retrieval
- Topic Maps
- Ontology
- Information Organization
Date of Defense 2006-03-31 Availability unrestricted Abstract Enhancing the representation and relationship of information through ontology is a promising alternative approach for knowledge organization. This improved knowledge organization is vital for collocation of information and effective and efficient searching. This study concerned the testing of user performance when searching an ontology-driven information retrieval (ONTOIR) system that shows explicit relationships among resources. The study explores the possibilities of improving user performance in searching for information. The goal was to examine whether or not ontology enhances user performance in terms of recall and search time. The experiment was conducted with 40 participants to evaluate and compare the differences in user performance (recall and search time) between an ontology-driven information retrieval system and a traditional, thesaurus-driven information retrieval system.
Better recall and shorter search time were found when conducting relationship-based queries in an ontology-driven information retrieval system as compared to a thesaurus-based system. . Further studies comparing user performance with a cluster-based search engine and an ontology-driven information retrieval system are needed.
Files
Filename Size Approximate Download Time (Hours:Minutes:Seconds)
28.8 Modem 56K Modem ISDN (64 Kb) ISDN (128 Kb) Higher-speed Access Myongho_Yi.pdf 2.90 Mb 00:13:25 00:06:54 00:06:02 00:03:01 00:00:15