규칙의 일반화와 통계 방식을 이용한 한국어 문맥의존 철자오류 교정 규칙의 재현율 향상 [韩语论文]

资料分类免费韩语论文 责任编辑:金一助教更新时间:2017-04-27
提示:本资料为网络收集免费论文,存在不完整性。建议下载本站其它完整的收费论文。使用可通过查重系统的论文,才是您毕业的保障。

Context-sensitive spelling errors (CSSE) are hard to correct, since they are perfect words when analyzed alone. Determined only by considering the semantic and syntactic relations of their context, CSSEs affect largely the performance of spelling and ...

Context-sensitive spelling errors (CSSE) are hard to correct, since they are perfect words when analyzed alone. Determined only by considering the semantic and syntactic relations of their context, CSSEs affect largely the performance of spelling and grammar checkers. The existing Korean Spelling and Grammar Checker (KSGC) adopts a rule-based method, which uses hand-made correction rules for CSSEs. Focusing on general users' demands, the KSGC is designed to obtain the very high precision, which results in the extremely low recall. This system does not meet the needs of text experts such as editors, writers or journalists, who wish to detect and correct as many errors as possible.

In order to improve the recall without lowering the precision of KSGC, this study proposes a method to expand conditions for selectional restrictions of argument using Korean Word to find the common superordinate conception. Also, this study proposes a generalize constraints on case marker(s) of an argument in existing error correction rules. But it is hard to improve the recall using only rule-based method. So this integrate the rule-based method and our previous works that correct CSSE using conditional probability.

韩语论文网站韩语毕业论文
免费论文题目: