|
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:
https://hdl.handle.net/10119/3909
|
| Title: | Sentence Extraction with Support Vector Machine Ensemble |
| Authors: | Minh, Le Nguyen
Shimazu, Akira
Xuan, Hieu Phan
Tu, Bao Ho
Horiguchi, Susumu |
| Keywords: | Text summarization
sentence extraction
SVM
Ensemble learning
SVM ensemble |
| Issue Date: | Nov-2005 |
| Publisher: | JAIST Press |
| Abstract: | This paper addresses a support vector machine model for text summarization problem. First, we formulate the text summarization problem as the problem of extracting a set of importance sentences. We then employ a support vector model for sloving that problem. Although the SVM are shown to be very suitable for sentence extraction because of the abillty in dealing with a very large of feature demision. The limitation of it is that in practical some approxiamtion algorithm are used. It might reduce the accuracy of classification. To overcome the above drawback, a SVM ensemble is clearly sutiable. This was because when combining each individual SVM has been traiend independently from the random chosen training samples and the correctly classifed area in the sapce of data samples of each SVM becomes limited to a certain area. We can expect that a combination of several SVMs will exapand the correctly classified area incrementlly. This paper initialy presents the use of ensemble SVM to text summarization and shows that the performance of SVM ensemble will be better than that of convential SVM. |
| 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, 2119, Kobe, Japan
Symposium 5, Session 2 : Data/Text Mining from Large Databases Text Mining |
| Language: | ENG |
| URI: | https://hdl.handle.net/10119/3909 |
| ISBN: | 4-903092-02-X |
| Appears in Collections: | IFSR 2005
|
Files in This Item:
| File |
Description |
Size | Format |
|---|
| 20073.pdf | | 163Kb | Adobe PDF | View/Open |
|
All items in DSpace are protected by copyright, with all rights reserved.
|
