|
JAIST Repository >
School of Information Science >
Articles >
Journal Articles >
Please use this identifier to cite or link to this item:
https://hdl.handle.net/10119/7832
|
| Title: | Automatic Extraction of the Fine Category of Person Named Entities from Text Corpora |
| Authors: | NGUYEN, Tri-Thanh
SHIMAZU, Akira |
| Keywords: | fine person categories extraction
named entities
pattern extraction
algorithm |
| Issue Date: | 2007-10-01 |
| Publisher: | 電子情報通信学会 |
| Magazine name: | IEICE TRANSACTIONS on Information and Systems |
| Volume: | E90-D |
| Number: | 10 |
| Start page: | 1542 |
| End page: | 1549 |
| DOI: | 10.1093/ietisy/e90-d.10.1542 |
| Abstract: | Named entities play an important role in many Natural Language Processing applications. Currently, most named entity recognition systems rely on a small set of general named entity (NE) types. Though some efforts have been proposed to expand the hierarchy of NE types, there are still a fixed number of NE types. In real applications, such as question answering or semantic search systems, users may be interested in more diverse specific NE types. This paper proposes a method to extract categories of person named entities from text documents. Based on Dual Iterative Pattern Relation Extraction method, we develop a more suitable model for solving our problem, and explore the generation of different pattern types. A method for validating whether a category is valid or not is proposed to improve the performance, and experiments on Wall Street Journal corpus give promising results. |
| Rights: | Copyright (C)2007 IEICE. Tri-Thanh Nguyen, Akira Shimazu, IEICE TRANSACTIONS on Information and Systems, E90-D(10), 2007, 1542-1549. http://www.ieice.org/jpn/trans_online/ |
| URI: | https://hdl.handle.net/10119/7832 |
| Material Type: | publisher |
| Appears in Collections: | b10-1. 雑誌掲載論文 (Journal Articles)
|
Files in This Item:
| File |
Description |
Size | Format |
|---|
| A11970.pdf | | 430Kb | Adobe PDF | View/Open |
|
All items in DSpace are protected by copyright, with all rights reserved.
|
