TY - CONF T1 - Linking Ontological Resources Using Aggregatable Substance Identifiers to Organize Extracted Relations T2 - Proceedings of the Pacific Symposium on Biocomputing, Jan 4-8, 2005, Big Island, Hawaii Y1 - 2005 A1 - Marshall,Byron A1 - Su,Hua A1 - McDonald,Dan A1 - Chen,Hsinchun KW - Accounting KW - BIS AB - Systems that extract biological regulatory pathway relations from free-text sources are
intended to help researchers leverage vast and growing collections of research literature.
Several systems to extract such relations have been developed but little work has focused on
how those relations can be usefully organized (aggregated) to support visualization systems or
analysis algorithms. Ontological resources that enumerate name strings for different types of
biomedical objects should play a key role in the organization process. In this paper we
delineate five potentially useful levels of relational granularity and propose the use of
aggregatable substance identifiers to help reduce lexical ambiguity. An aggregatable
substance identifier applies to a gene and its products. We merged 4 extensive lexicons and
compared the extracted strings to the text of five million MEDLINE abstracts. We report on
the ambiguity within and between name strings and common English words. Our results show
an 89% reduction in ambiguity for the extracted human substance name strings when using an
aggregatable substance approach. JA - Proceedings of the Pacific Symposium on Biocomputing, Jan 4-8, 2005, Big Island, Hawaii UR - http://people.oregonstate.edu/~marshaby/Papers/marshall_PSB2005.pdf U2 - b U4 - 2606753793 ID - 2606753793 ER - TY - JOUR T1 - EBizPort: Collecting and Analyzing Business Intelligence Information JF - Journal of the Association for Information Science and Technology Y1 - 2004 A1 - Marshall,Byron A1 - McDonald,Dan A1 - Chen,Hsinchun A1 - Chung,Wingyan KW - Accounting KW - BIS AB - In this article, Marshall, McDonald, Chen, and Chung take a different approach to supporting search services to large and heterogeneous document collections. They propose development of a domain-specific collection by crawling the content of a small set of highly reputable sites, maintaining a local index of the content, and providing browsing and searching services on the specialized content. This resource, known as a vertical portal, has the potential of overcoming several problems associated with bias, update delay, reputation, and integration of scattered information. The article discusses the design of a vertical portal system's architecture called EbizPort, rationale behind its major components, and algorithms and techniques for building collections and search functions. Collection (or more broadly content) has an obvious relationship to the nature of the search interface, as it can impact the type of search functions that can be offered. Powerful search interface functions were built for EbizPort by exploiting the underlying content representation and a relatively narrow and well-defined domain focus. Particularly noteworthy are the innovative browsing functions, which include a summarizer, a categorizer, a visualizer, and a navigation side-bar. The article ends with a discussion of an evaluation study, which compared the EbizPort system with a baseline system called Brint. Results are presented on effectiveness and efficiency, usability and information quality, and quality of local collection and content retrieved from other sources (an extended search operation called meta-search service was also provided in the system). Overall, the authors find that EbizPort outperforms the baseline system, and it provides a viable way to support access to business information. VL - 55 UR - http://people.oregonstate.edu/~marshaby/Papers/Marshall_JASIST_EBizPort.pdf CP - 1 U2 - a U4 - 648210432 ID - 648210432 ER - TY - JOUR T1 - Extracting Gene Pathway Relations Using a Hybrid Grammar: The Arizona Relation Parser JF - Bioinformatics Y1 - 2004 A1 - McDonald,Dan A1 - Chen,Hsinchun A1 - Su,Hua A1 - Marshall,Byron KW - Accounting KW - BIS AB - Motivation: Text-mining research in the biomedical domain has been motivated by the rapid growth of new research findings. Improving the accessibility of findings has potential to speed hypothesis generation.Results: We present the Arizona Relation Parser that differs from other parsers in its use of a broad coverage syntax-semantic hybrid grammar. While syntax grammars have generally been tested over more documents, semantic grammars have outperformed them in precision and recall. We combined access to syntax and semantic information from a single grammar. The parser was trained using 40 PubMed abstracts and then tested using 100 unseen abstracts, half for precision and half for recall. Expert evaluation showed that the parser extracted biologically relevant relations with 89% precision. Recall of expert identified relations with semantic filtering was 35 and 61% before semantic filtering. Such results approach the higher-performing semantic parsers. However, the AZ parser was tested over a greater variety of writing styles and semantic content. VL - 20 UR - http://people.oregonstate.edu/~marshaby/Papers/MCDONALD_BIOINFORMATICS.pdf CP - 18 U2 - a U4 - 648206336 ID - 648206336 ER -