Type of Document Dissertation Author Li, Xin URN etd-04052006-145601 Title INCREMENTAL QUERY PROCESSING IN INFORMATION FUSION SYSTEMS Degree Doctor of Philosophy Program Computer Science School School of Arts and Sciences Advisory Committee
Advisor Name Title Shi-Kuo Chang Committee Chair Alexandros Labrinidis Committee Member Ching-Chung Li Committee Member Janyce Wiebe Committee Member Keywords
- Information Fusion
- Query Pattern
- Incremental Query
- Query Processing
Date of Defense 2006-02-23 Availability unrestricted Abstract This dissertation studies the methodology and techniques of information retrieval in fusion systems where information referring to same objects is assessed on the basis of data from multiple heterogeneous data sources. A wide range of important applications can be categorized as information fusion systems e.g. multisensor surveillance system, local search system, multisource medical diagnose system, and so on. Up to the time of this dissertation, most information retrieval methods in fusion systems are highly domain specific, and most query systems do not address fusion problem with enough efforts.
In this dissertation, I describe a broadly applicable query based information retrieval approach in general fusion systems: user information needs are interpreted as fusion queries, and the query processing techniques e.g. source dependence graph (SDG), query refinement and optimization are described. Aiming to remove the query building bottleneck, a novel incremental query method is proposed, which can eliminate the accumulated complexity in query building as well as in query execution. Query pattern is defined to capture and reuse repeated structures in the incremental queries. Several new techniques for query pattern matching and learning are described in detail. Some important experiments in a real-world multisensor fusion system, i.e. the intelligent vehicle tracking (IVET) system, have been presented to validate the proposed methodology and techniques.
Files
Filename Size Approximate Download Time (Hours:Minutes:Seconds)
28.8 Modem 56K Modem ISDN (64 Kb) ISDN (128 Kb) Higher-speed Access LiXin04052006.pdf 2.33 Mb 00:10:47 00:05:33 00:04:51 00:02:25 00:00:12 If you have questions or comments please send mail to ETD-Feedback.