JAIST Repository >
Center for Strategic Development of Science and Technology 2003-2008 >
JAIST PRESS Publications >
IFSR 2005 >

Please use this identifier to cite or link to this item: http://hdl.handle.net/10119/3916

Title: Extracting Background Knowledge from the Medical Literature
Authors: Kawasaki, Saori
Tu, Bao Ho
Keywords: text mining
rough sets
MEDLINE
Issue Date: Nov-2005
Publisher: JAIST Press
Abstract: The growing interest in dealing with the information in medical publications is not only because of the scene of physicians’ medical practices as the evidence-based medicine (EBM) but also encouraged by the realization of computational approaches to the electrically stored medical data, such as data mining and many attempts to enrich them. Among many different purposes of means of utilizing MEDLINE, our target is to extract background knowledge from MEDLINE abstracts, and exploit it in the mining from a certain medical database. Generally, information extraction approaches try to understand the contents in text, and extract the appropriate information to the requests. While those requires complex and time consuming procedures to complete, we propose a method for extracting background knowledge useful for data mining from medical database in a simple but effective manner. Main idea of this method is to find strong combinations among clinical test items in order to use the result for feature selection and narrowing the mined result. We design the framework based on rough set theory for considering the word ambiguity. This framework consists of three steps: to approximately represent MEDLINE abstracts related to the topic under investigation by the tolerance rough set model, then to detect associations between terms related to the topic, and finally to discover rules with our rule learning program LUPC when considering found associations as input to exclusive or inclusive constraints of LUPC. Some sets of parameters succeed to extract plausibly useful combinations of clinical test items for finding patterns from the hepatitis database.
Description: The original publication is available at JAIST Press http://www.jaist.ac.jp/library/jaist-press/index.html
IFSR 2005 : Proceedings of the First World Congress of the International Federation for Systems Research : The New Roles of Systems Sciences For a Knowledge-based Society : Nov. 14-17, 2126, Kobe, Japan
Symposium 5, Session 4 : Data/Text Mining from Large Databases Text Mining
Language: ENG
URI: http://hdl.handle.net/10119/3916
ISBN: 4-903092-02-X
Appears in Collections:IFSR 2005

Files in This Item:

File Description SizeFormat
20177.pdf402KbAdobe PDFView/Open

All items in DSpace are protected by copyright, with all rights reserved.

 


Contact : Library Information Section, Japan Advanced Institute of Science and Technology