Date of Award
9-20-2010
Document Type
Dissertation
Degree Name
Doctor of Philosophy (PhD)
Department
Applied Science
First Advisor
Srinivasan Ramaswamy
Abstract
Signal detection is a challenging task for regulatory and intelligence agencies. Subject matter experts in those agencies analyze documents, generally containing narrative text in a time bound manner for signals by identification, evaluation and confirmation, leading to follow-up action e.g., recalling a defective product or public advisory for potential flu outbreak. Technical challenges to achieve effective signal detection include mining unstructured data, increasing document collections, and presence of multi-domain vocabulary. Lack of annotation and multi-domain vocabulary makes traditional semantic web mining ineffective. Use of heuristics to enhance bag-of-words type text mining to semantic text mining is a novel idea attempted for signal detection in this thesis. A semantic text mining framework based on information retrieval and extraction techniques for signal detection is presented as a solution. The framework has been validated by analyzing narrative text in medical device event reports and related documents available to the United States Food and Drug Administration.
Recommended Citation
Sudarsan, Sithu D., "Signal Detection Framework Using Semantic Text Mining Techniques" (2010). Theses and Dissertations. 246.
https://research.ualr.edu/etd/246
