The number of published literature documents is growing enormously. While growth in the biosciences field is encouraging for researchers, it also brings the problems of information overload, making it difficult to keep abreast of all of the developments in the field. Information retrieval systems (search engines) alone would not be sufficient to address the needs of researchers. Information extraction (IE) tools, which extract specific pieces of information from text documents for the user queries, are extremely valuable.
In this dissertation, we propose a structure called bio-semantic token subsequence to capture semantic features from natural language sentences for biomedical relationship extraction. We developed two supervised learning algorithms for extracting semantic relationships between entities. The first approach "bio-semantic token subsequence kernel" implicitly utilizes features captured by the bio-semantic token subsequences. The second learning approach is called "discriminative bio-semantic token subsequence classifier", which explicitly generates a discriminative subset of bio-semantic token subsequences. In our experimental evaluations, both of these proposed methods outperformed the state-of-the-art methods reported in the literature.
A key issue in text mining is the linking together of the extracted information to form new facts or hypotheses, which can subsequently be explored further by conventional means of experimentation. In this dissertation, we modeled biomedical literature repository as a comprehensive network of biomedical concepts and proposed and evaluated a supervised link discovery method for performing large-scale cross-silo biomedical hypothesis discovery.
|Advisor:||Raghavan, Vijay V.|
|Commitee:||Chu, Chee-Hung Henry, Efe, Kemal|
|School:||University of Louisiana at Lafayette|
|School Location:||United States -- Louisiana|
|Source:||DAI-B 74/05(E), Dissertation Abstracts International|
|Keywords:||Bioinformatics, Data mining, Information extraction, Link discovery, Machine learning, Nlp, Text mining|
Copyright in each Dissertation and Thesis is retained by the author. All Rights Reserved
The supplemental file or files you are about to download were provided to ProQuest by the author as part of a
dissertation or thesis. The supplemental files are provided "AS IS" without warranty. ProQuest is not responsible for the
content, format or impact on the supplemental file(s) on our system. in some cases, the file type may be unknown or
may be a .exe file. We recommend caution as you open such files.
Copyright of the original materials contained in the supplemental file is retained by the author and your access to the
supplemental files is subject to the ProQuest Terms and Conditions of use.
Depending on the size of the file(s) you are downloading, the system may take some time to download them. Please be