Kana-Kanji Conversion by A Stochastic Model
-
- MORI Shinsuke
- Tokyo Research Laboratory, IBM Japan
-
- TSUCHIYA Masatoshi
- Graduate School of Informatics, Kyoto University
-
- YAMAJI Osamu
- Graduate School of Informatics, Kyoto University
-
- NAGAO Makoto
- Graduate School of Informatics, Kyoto University
Bibliographic Information
- Other Title
-
- 確率的モデルによる仮名漢字変換
Search this article
Abstract
In this paper, we present a kana-kanji converter by a stochastic model. Given an input this method returns the most probable kana-kanji character sequence. For its evaluation, we converted kana sequences of a corpus containing kana-kanji sequences. The criterion we used is the ratio of the length of longest common subsequece. The recall and precision of our method are 95.07% and 93.94% respectively. This result is much better than that of Wnn6 (recall : 91.12% ; precision : 91.17%).
Journal
-
- IPSJ SIG Notes
-
IPSJ SIG Notes 125 93-99, 1998-05-28
Information Processing Society of Japan (IPSJ)
- Tweet
Details 詳細情報について
-
- CRID
- 1572543027084199936
-
- NII Article ID
- 110002934730
-
- NII Book ID
- AN10115061
-
- Text Lang
- ja
-
- Data Source
-
- CiNii Articles