|
JAIST Repository >
School of Knowledge Science >
Articles >
Journal Articles >
Please use this identifier to cite or link to this item:
http://hdl.handle.net/10119/5002
|
| Title: | Combining classifiers for word sense disambiguation based on Dempster-Shafer theory and OWA operators |
| Authors: | Le, Anh Cuong
Huynh, Van-Nam
Shimazu, Akira
Nakamori, Yoshiteru |
| Keywords: | Computational linguistics
Classifier combination
Word sense disambiguation
OWA operator
Evidential reasoning |
| Issue Date: | 2007-11 |
| Publisher: | Elsevier |
| Magazine name: | Data & Knowledge Engineering |
| Volume: | 63 |
| Number: | 2 |
| Start page: | 381 |
| End page: | 396 |
| DOI: | 10.1016/j.datak.2007.03.013 |
| Abstract: | In this paper, we discuss a framework for weighted combination of classifiers for word sense disambiguation (WSD). This framework is essentially based on Dempster-Shafer theory of evidence (Shafer, 1976) and ordered weighted averaging (OWA) operators (Yager, 1988). We first determine various kinds of features which could provide complementarily linguistic information for the context, and then combine these sources of information based on Dempster's rule of combination and OWA operators for identifying the meaning of a polysemous word. We experimentally design a set of individual classifiers, each of which corresponds to a distinct representation type of context considered in the WSD literature, and then the discussed combi-nation strategies are tested and compared on English lexical samples of Senseval-2 and Senseval-3. |
| Rights: | NOTICE: This is the author's version of a work accepted for publication by Elsevier. Cuong Anh Le, Van-Nam Huynh, Akira Shimazu and Yoshiteru Nakamori, Data & Knowledge Engineering, 63(2), 2007, 381-396, http://dx.doi.org/10.1016/j.datak.2007.03.013 |
| URI: | http://hdl.handle.net/10119/5002 |
| Material Type: | author |
| Appears in Collections: | a10-1. 雑誌掲載論文 (Journal Articles)
|
Files in This Item:
| File |
Description |
Size | Format |
|---|
| DKE-final.pdf | | 243Kb | Adobe PDF | View/Open |
|
All items in DSpace are protected by copyright, with all rights reserved.
|
