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

Continuous analysis of Internet text...

Jorgensen, Peter Earl.

FindBook

Google Book

Amazon

博客來

Continuous analysis of Internet text by artificial neural network.

紀錄類型:	書目-語言資料,印刷品 : Monograph/item
正題名/作者:	Continuous analysis of Internet text by artificial neural network./
作者:	Jorgensen, Peter Earl.
面頁冊數:	73 p.
附註:	Adviser: Joseph Woelfel.
Contained By:	Dissertation Abstracts International63-12A.
標題:	Artificial Intelligence. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3076493
ISBN:	0493968458

Continuous analysis of Internet text by artificial neural network.
Jorgensen, Peter Earl.

Continuous analysis of Internet text by artificial neural network. - 73 p.

Adviser: Joseph Woelfel.

Thesis (Ph.D.)--State University of New York at Buffalo, 2003.

Many segments of modern society desire the ability to find relationships, thus meaning, in public discourse. Notable segments include marketing, politics, government, activism and public safety. The increasing use of the Internet for public dialog has made Internet communication a potentially rich source of information in this regard. This study explores the use of an Interactive Activation with Competition (IAC) artificial neural network to aid in processing text to find relationships. A fully recurrent IAC network was modified to process text in a continuous fashion, adding nodes as new terms were encountered. Forty-nine email messages from two threads in the Open Library/Information Science Education Forum were processed using two variations in the self-organizing phase of network formation. The messages were processed with and without a linear decay function applied to the external activation of nodes between sentences. With the function self-organization includes reduced external activation levels of terms in sentences that have already been processed in the current message. Without it self-organization externally activates only terms in the sentence currently being processed. This could be considered a type of context control. The use of the linear decay function produced three effects. When the function was used, roughly half the number of noise strings were highly associated with key terms; the entire network was more differentiated from key terms and; the key terms were more highly associated with each other. These effects could reduce or eliminate the need for stop word filtering as well as improve system performance. Future research should explore refinements in this concept, as well as its ability to disambiguate terms. Combining this approach with models of discourse structure is another area for research.

ISBN: 0493968458Subjects--Topical Terms:

769149
Artificial Intelligence.

Continuous analysis of Internet text by artificial neural network.
LDR:02753nam 2200289 a 45 001 928303
005 20110426
008 110426s2003 eng d
020 $a 0493968458
035 $a (UnM)AAI3076493
035 $a AAI3076493
040 $a UnM $c UnM
100 1 $a Jorgensen, Peter Earl. $3 1251764
245 1 0 $a Continuous analysis of Internet text by artificial neural network.
300 $a 73 p.
500 $a Adviser: Joseph Woelfel.
500 $a Source: Dissertation Abstracts International, Volume: 63-12, Section: A, page: 4137.
502 $a Thesis (Ph.D.)--State University of New York at Buffalo, 2003.
520 $a Many segments of modern society desire the ability to find relationships, thus meaning, in public discourse. Notable segments include marketing, politics, government, activism and public safety. The increasing use of the Internet for public dialog has made Internet communication a potentially rich source of information in this regard. This study explores the use of an Interactive Activation with Competition (IAC) artificial neural network to aid in processing text to find relationships. A fully recurrent IAC network was modified to process text in a continuous fashion, adding nodes as new terms were encountered. Forty-nine email messages from two threads in the Open Library/Information Science Education Forum were processed using two variations in the self-organizing phase of network formation. The messages were processed with and without a linear decay function applied to the external activation of nodes between sentences. With the function self-organization includes reduced external activation levels of terms in sentences that have already been processed in the current message. Without it self-organization externally activates only terms in the sentence currently being processed. This could be considered a type of context control. The use of the linear decay function produced three effects. When the function was used, roughly half the number of noise strings were highly associated with key terms; the entire network was more differentiated from key terms and; the key terms were more highly associated with each other. These effects could reduce or eliminate the need for stop word filtering as well as improve system performance. Future research should explore refinements in this concept, as well as its ability to disambiguate terms. Combining this approach with models of discourse structure is another area for research.
590 $a School code: 0656.
650 4 $a Artificial Intelligence. $3 769149
650 4 $a Information Science. $3 1017528
650 4 $a Library Science. $3 881164
690 $a 0399
690 $a 0723
690 $a 0800
710 2 0 $a State University of New York at Buffalo. $3 1017814
773 0 $t Dissertation Abstracts International $g 63-12A.
790 $a 0656
790 1 0 $a Woelfel, Joseph, $e advisor
791 $a Ph.D.
792 $a 2003
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3076493