東華大學圖書館 |

語系: 繁體中文

說明(常見問題)

回圖書館首頁

手機版館藏查詢

登入

回首頁

切換: 標籤 | MARC模式 | ISBD

Text mining in GeneRIFs.

University of Colorado Health Sciences Center.

FindBook

Google Book

Amazon

博客來

Text mining in GeneRIFs.

紀錄類型:	書目-語言資料,印刷品 : Monograph/item
正題名/作者:	Text mining in GeneRIFs./
作者:	Lu, Zhiyong.
面頁冊數:	182 p.
附註:	Adviser: Lawrence Hunter.
Contained By:	Dissertation Abstracts International68-05B.
標題:	Biology, Bioinformatics. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3264544
ISBN:	9780549033301

Text mining in GeneRIFs.
Lu, Zhiyong.

Text mining in GeneRIFs. - 182 p.

Adviser: Lawrence Hunter.

Thesis (Ph.D.)--University of Colorado Health Sciences Center, 2007.

With the double-exponential growth of the peer-reviewed literature, the amount of information relevant to an experimental biologist is exploding, thus making it harder than ever for an individual to find and assimilate all relevant knowledge from the literature. An alternative approach to manual inspection of the literature is to employ computational methods to process textual inputs such as abstracts and full-text articles. This approach is known as text mining. In this thesis I develop a text-mining framework for extracting protein transport information from Gene References Into Function (GeneRIFs). GeneRIFs are short descriptions about the function of genes derived from the primary literature. I will show that this framework provides state-of-the-art accuracy, and is generalizable to other text-mining tasks.

ISBN: 9780549033301Subjects--Topical Terms:

1018415
Biology, Bioinformatics.

Text mining in GeneRIFs.
LDR:03434nam 2200289 a 45 001 861644
005 20100720
008 100720s2007 ||||||||||||||||| ||eng d
020 $a 9780549033301
035 $a (UMI)AAI3264544
035 $a AAI3264544
040 $a UMI $c UMI
100 1 $a Lu, Zhiyong. $3 1029362
245 1 0 $a Text mining in GeneRIFs.
300 $a 182 p.
500 $a Adviser: Lawrence Hunter.
500 $a Source: Dissertation Abstracts International, Volume: 68-05, Section: B, page: 2742.
502 $a Thesis (Ph.D.)--University of Colorado Health Sciences Center, 2007.
520 $a With the double-exponential growth of the peer-reviewed literature, the amount of information relevant to an experimental biologist is exploding, thus making it harder than ever for an individual to find and assimilate all relevant knowledge from the literature. An alternative approach to manual inspection of the literature is to employ computational methods to process textual inputs such as abstracts and full-text articles. This approach is known as text mining. In this thesis I develop a text-mining framework for extracting protein transport information from Gene References Into Function (GeneRIFs). GeneRIFs are short descriptions about the function of genes derived from the primary literature. I will show that this framework provides state-of-the-art accuracy, and is generalizable to other text-mining tasks.
520 $a There are four major components within the framework. The first component aims to identify biological entities (gene/proteins and cellular components) that are relevant to protein transport in GeneRIFs and to associate the identified entities to standard database identifiers (Entrez Gene IDs for gene/proteins and Gene Ontology IDs for cellular components). The second component---the core of the work---is an ontology-driven system (called Open Source Direct Memory Access Parser, or OpenDMAP) for characterizing complex relationships among those entities (e.g., which protein gets transported where). The third component of the framework involves the construction of a gold-standard corpus of protein transport in order to evaluate the first two components. The last component consists of two systems that find new GeneRIFs and inspect the quality of existing GeneRIFs respectively, thus guaranteeing a sufficient number of high-quality textual inputs for the analytical framework.
520 $a The entire framework is described, and the results of its application are presented, along with evaluative experiments. The framework obtained excellent results for gene normalization, transport information extraction, GeneRIF prediction, and GeneRIF quality control. An additional third-party evaluation was performed by applying the framework to three other tasks in the context of the BioCreAtIvE 2006 shared tasks without major changes. The preliminary results from the BioCreAtIvE organizers show that our performance is better than published results. Thus, I will argue that the framework is generalizable and applicable to other text-mining tasks. In summary, this dissertation demonstrates an accurate and extensible framework for text mining on GeneRIFs.
590 $a School code: 0831.
650 4 $a Biology, Bioinformatics. $3 1018415
690 $a 0715
710 2 $a University of Colorado Health Sciences Center. $3 1023815
773 0 $t Dissertation Abstracts International $g 68-05B.
790 $a 0831
790 1 0 $a Hunter, Lawrence, $e advisor
791 $a Ph.D.
792 $a 2007
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3264544