東華大學圖書館 |

Language: English

Help

回圖書館首頁

手機版館藏查詢

Back

Switch To: Labeled | MARC Mode | ISBD

Categorization of electronic medical...

Zheng, Wanhong.

Linked to FindBook

Google Book

Amazon

博客來

Categorization of electronic medical news.

Record Type:	Language materials, printed : Monograph/item
Title/Author:	Categorization of electronic medical news./
Author:	Zheng, Wanhong.
Description:	125 p.
Notes:	Advisers: Carolyn R. Watters; Evangelos E. Milios.
Contained By:	Masters Abstracts International40-05.
Subject:	Computer Science. -
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=MQ66610
ISBN:	0612666107

Categorization of electronic medical news.
Zheng, Wanhong.

Categorization of electronic medical news. - 125 p.

Advisers: Carolyn R. Watters; Evangelos E. Milios.

Thesis (M.Comp.Sc.)--Dalhousie University (Canada), 2001.

This thesis presents a keyword-based medical newspaper article classification system. The system identifies the keywords that characterize the medical content of the article and classifies the article into four classes: non-medical, medical for layperson, medical for medicine-knowledgeable and medical for expert. Additionally, a classified medical article can be tagged with MeSH concepts for the purpose of content description.

ISBN: 0612666107Subjects--Topical Terms:

626642
Computer Science.

Categorization of electronic medical news.
LDR:02118nam 2200325 a 45 001 928444
005 20110426
008 110426s2001 eng d
020 $a 0612666107
035 $a (UnM)AAIMQ66610
035 $a AAIMQ66610
040 $a UnM $c UnM
100 1 $a Zheng, Wanhong. $3 1251905
245 1 0 $a Categorization of electronic medical news.
300 $a 125 p.
500 $a Advisers: Carolyn R. Watters; Evangelos E. Milios.
500 $a Source: Masters Abstracts International, Volume: 40-05, page: 1264.
502 $a Thesis (M.Comp.Sc.)--Dalhousie University (Canada), 2001.
520 $a This thesis presents a keyword-based medical newspaper article classification system. The system identifies the keywords that characterize the medical content of the article and classifies the article into four classes: non-medical, medical for layperson, medical for medicine-knowledgeable and medical for expert. Additionally, a classified medical article can be tagged with MeSH concepts for the purpose of content description.
520 $a In categorization core of the system, MeSH (Medical Subject Headings) hierarchical vocabulary is used as both a medical dictionary and categorization taxonomy. A manual pre-defined weighting scheme on MeSH concepts is used to differentiate medical keywords of different “difficulty” levels. Handcrafted classification algorithm has been investigated and the overall correctness is around 75%. Machine learning approach has been explored to construct 3-way classification rules automatically. While further tuning of medical training articles is proved to be necessary in order to get an effective classifier, the correctness on finding non-medical articles for our learning classifier is shown to be 92%.
590 $a School code: 0328.
650 4 $a Computer Science. $3 626642
650 4 $a Health Sciences, General. $3 1017817
650 4 $a Information Science. $3 1017528
650 4 $a Journalism. $3 576107
690 $a 0391
690 $a 0566
690 $a 0723
690 $a 0984
710 2 0 $a Dalhousie University (Canada). $3 1017625
773 0 $t Masters Abstracts International $g 40-05.
790 $a 0328
790 1 0 $a Milios, Evangelos E., $e advisor
790 1 0 $a Watters, Carolyn R., $e advisor
791 $a M.Comp.Sc.
792 $a 2001
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=MQ66610