• Proceedings of the Korean Society for Cognitive Science Conference
• /
• 2002.05a
• /
• pp.46-50
• /
• 2002

인간의 지식 표상 규명에 대한 연구는 인간을 대상으로 연구하는 심리학에서뿐만 아니라 인간의 지능을 컴퓨터를 이용하여 구현하고자 하는 인공지능 학문에서도 오래 전부터 매우 중요한 화두가 되고 있다. 특히 인간의 지식 중 언어 지식에 대한 연구는 인간의 언어처리 과정 및 현상을 규명하고 이해하고자 하는 심리언어학에서뿐만 아니라 인간의 언어를 컴퓨터를 이용하여 처리하고자 하는 전산언어학 연구에 있어서도 매우 중요하다. 본 논문은 피험자를 대상으로 한 어절 재인 시 관찰되었던 언어 현상을 설명할 수 있는 시뮬레이션 모델과 이에 근거한 심성어휘집내에서의 한국어 어절의 표상 구조를 제안한다.

• Journal of KIISE
• /
• v.44 no.10
• /
• pp.1087-1093
• /
• 2017

Educating foreigners in Korean language is attracting increasing attention with the growing number of foreigners who want to learn Korean or want to reside in Korea. Existing spell checkers mostly focus on native Korean speakers, so they are inappropriate for foreigners. In this paper, we propose a correction method for the Korean language that reflects the contextual characteristics of Korean and writing characteristics of foreigners. Our method can extract frequently used expressions by Koreans by constructing syllable reverse-index for eojeol bi-gram extracted from corpus as correction candidates, and generate ranked Korean corrections for foreigners with upgraded edit distance calculation. Our system provides a user interface based on keyboard hooking, so a user can easily use the correction system along with other applications. Our system improves the detection rate for foreign language users by about 45% compared to other systems in foreign language writing environments. This will help foreign users to judge and correct their own writing errors.

• Journal of KIISE
• /
• v.43 no.3
• /
• pp.362-369
• /
• 2016

This study is directed toward the design of a hybrid algorithm for syllable-based Korean POS tagging. Previous syllable-based works on Korean POS tagging have relied on a sequence labeling method and mostly used only a machine learning method. We present a new algorithm integrating a machine learning method and a pre-analyzed dictionary. We used a Sejong tagged corpus for training and evaluation. While the machine learning engine achieved eojeol precision of 0.964, the proposed hybrid engine achieved eojeol precision of 0.990. In a Quiz domain test, the machine learning engine and the proposed hybrid engine obtained 0.961 and 0.972, respectively. This result indicates our method to be effective for Korean POS tagging.

• Korean Journal of Cognitive Science
• /
• v.20 no.3
• /
• pp.291-307
• /
• 2009

This study investigated global and local characteristics of eye movement while 30 college students read easy and difficult Korean texts. It was found that readers who read the difficult text fixated longer for about 217ms and made shorter saccades of about 3.7 characters while readers who read the easy one fixated for about 190ms and made saccades of about 4.8 characters. Single fixation times and gaze durations in the difficult text were longer than those in the easy one(227ms vs. 195ms; 266ms vs. 210ms). In both easy and difficult texts, the effects of word frequency and eojeol length were found. In addition, the differences in fixation times according to word frequency and length were larger in the difficult text.

• Journal of KIISE:Software and Applications
• /
• v.30 no.11
• /
• pp.1112-1123
• /
• 2003

WSD(word sense disambiguation) is one of the most difficult problems in Korean information processing. The Bayesian model that used semantic information, extracted from definition corpus(1 million POS-tagged eojeol, Korean dictionary definitions), resulted in accuracy of 72.08% (nouns 78.12%, verbs 62.45%). This paper proposes the statistical WSD model using NPH(New Prior Probability of Homonym sense) and distance weights. We select 46 homonyms(30 nouns, 16 verbs) occurred high frequency in definition corpus, and then we experiment the model on 47,977 contexts from ‘21C Sejong Corpus’(3.5 million POS-tagged eojeol). The WSD model using NPH improves on accuracy to average 1.70% and the one using NPH and distance weights improves to 2.01%.

• Journal of KIISE
• /
• v.41 no.12
• /
• pp.1081-1089
• /
• 2014

Context-sensitive spelling-error correction methods are largely classified into rule-based methods and statistical data-based methods, the latter of which is often preferred in research. Statistical error correction methods consider context-sensitive spelling error problems as word-sense disambiguation problems. The method divides a vocabulary pair, for correction, which consists of a correction target vocabulary and a replacement candidate vocabulary, according to the context. The present paper proposes a method that integrates a word-phrase n-gram model into a conventional model in order to improve the performance of the probability model by using a correction vocabulary pair, which was a result of a previous study performed by this research team. The integrated model suggested in this paper includes a method used to interpolate the probability of a sentence calculated through each model and a method used to apply the models, when both methods are sequentially applied. Both aforementioned types of integrated models exhibit relatively high accuracy and reproducibility when compared to conventional models or to a model that uses only an n-gram.

• Phonetics and Speech Sciences
• /
• v.6 no.3
• /
• pp.155-164
• /
• 2014

This paper proposes a new method to determine the recognition units for large vocabulary continuous speech recognition (LVCSR) in Korean by applying unsupervised segmentation and merging. In the proposed method, a text sentence is segmented into morphemes and position information is added to morphemes. Then submorpheme units are obtained by splitting the morpheme units through the maximization of posterior probability terms. The posterior probability terms are computed from the morpheme frequency distribution, the morpheme length distribution, and the morpheme frequency-of-frequency distribution. Finally, the recognition units are obtained by sequentially merging the submorpheme pair with the highest frequency. Computer experiments are conducted using a Korean LVCSR with a 100k word vocabulary and a trigram language model obtained by a 300 million eojeol (word phrase) corpus. The proposed method is shown to reduce the out-of-vocabulary rate to 1.8% and reduce the syllable error rate relatively by 14.0%.

• Proceedings of the Korean Information Science Society Conference
• /
• 2008.06c
• /
• pp.279-284
• /
• 2008

본 연구에서는 어절패턴을 이용하는 새로운 방식의 한국어 형태소 분석기 KGuru-MA에 대해서 설명한다. KGuru-MA는 품사 부착 말뭉치에서 개방어를 생략하여 어절 패턴을 반자동으로 학습하여 어절 패턴 사전과 형태소 확률 정보 사전을 구성한 후, 이 사전을 이용하여 형태소를 분석한다. 본 형태소 분석기는 어절패턴을 사용하여 형태소 분석하기 때문에 기존 형태소 분석기에 존재하는 접속검사 과정이 생략된다. 또한, 형태소 분석 과정이 기존의 형태소 분석기에 비해 단순하여 기초 자연언어 처리 시스템이 가지는 강건성을 보장한다. 본 연구는 "21세기 세종기획 3차년도 말뭉치"를 이용한 실험 결과, 기존 형태소 분석기 못지 않은 성능을 보였다.

• Annual Conference on Human and Language Technology
• /
• 2019.10a
• /
• pp.43-48
• /
• 2019

본 논문에서는 한국어 챗봇에서의 문장 분류 시스템에 대하여 서술한다. 텍스트를 입력으로 받는 한국어 챗봇의 경우, 때때로 입력 문장에 오타나 띄어쓰기 오류 등이 포함될 수 있고, 이러한 오류는 잘못된 형태소 분석 결과로 이어지게 된다. 잘못된 형태소 분석 결과로 인한 문장 분류의 오류를 줄이기 위하여, 본 논문에서는 새로운 통합 어절 임베딩 방식을 제안한다. 통합 어절 임베딩 방식의 단점을 보완하고 성능을 향상시키기 위하여, 두 가지의 말뭉치 노이즈 추가 방법이 별도로 제안되었다. 실험 결과에 따르면, 본 논문에서 제안된 시스템은 오류를 포함한 한국어 문장 분류 문제에서 기존 시스템과 비교하여 문장 단위 정확률 기준으로 23 %p의 성능 향상을 보였다.

• Annual Conference on Human and Language Technology
• /
• 1997.10a
• /
• pp.327-331
• /
• 1997

본 논문에서는 기계번역과 의미분석의 전단계로서의 구문분석에 대하여 논한다. 의존 문법에 기반을 둔 구문분석의 효율성을 위하여 한국어 어절에 대한 새로운 해석을 시도하며, 이를 기반으로 한국어 의존관계 파서의 새로운 기본 단위(SynN: Syntactic Node)를 제시한다. 또한 새로운 기본 단위를 구문분석 과정에 적용하는 방법과 그 결과를 보인다. 마지막으로, 구현된 구문분석기를 중간언어 방식 시스템인 한-중 기계번역 시스템에 채용하여 그 성능을 검증한다.