• Korean Journal of Cognitive Science
• /
• v.10 no.2
• /
• pp.11-16
• /
• 1999

It is expected that some Josa/Eomi's are frequently used and others are not in the Korean documents. In this paper. we confirm it through the experiment and show that such information is very useful for Korean language processing. In case of Josa. most frequent 9 Josa's occupied 70% of total Josa's and 20. 32. 69 Josa's occupied 90%. 95%. and 99% respectively. Similarly, most frequent 10 numbers of Eomi's occupied 70% of total Eomi's and 33. 54. 117 Eomi's occupied 90%. 95%. and 99% respectively. We propose a dictionary construction method for Josa/Eomi dictionary that is classified by the frequency information. Furthermore. Josa/Eomi frequency results are very useful for the identification of unregistered morphemes and the disambiguation of lexical ambiguities.

• Annual Conference on Human and Language Technology
• /
• 2007.10a
• /
• pp.228-232
• /
• 2007

In this paper we propose comparison of both verb and noun endings between Korean and Mongolian languages. It is based on the similarity between two languages which have the same sentence structures and their eomi and josa structure. Korean verb and noun endings match into those of corresponding Mongolian endings. Josa and eomi are classified as a one-to-one, a one-to-many, and a many-to-many cases as well as some abnormal cases. In order to encourage development of Korean to Mongolian machine translation system, this paper would introduce one of the significant units in grammar.

• MALSORI
• /
• no.65
• /
• pp.125-141
• /
• 2008

In this work, we propose an automatic word spacing system for Korean using conditional random fields (CRFs) with Korean features. We map a word spacing problem into a classification problem in our work. We build a basic system which uses CRFs and Eumjeol bigram. After then, we analyze the result of inner-test. We extend a basic system added by some Korean features which are Josa, Eomi and two head Eumjeols of word extracting from lexicon. From the results of experiment, we can see that the proposed method is better than previous methods. Additionally the proposed method will be able to use mobile and speech applications because of very small size of model.

• Annual Conference on Human and Language Technology
• /
• 2001.10d
• /
• pp.97-100
• /
• 2001

본 논문은 규칙과 통계 정보를 모두 적용하는 혼합형 품사 태깅 시스템에서 통계 정보를 이용하여 품사 태깅을 수행할 때 조사와 어미를 문법 기능에 따라 구분하여 사용하는 품사 태깅 시스템을 기술한파. 품사 태깅은 주로 주변의 품사열을 이용하게 되는데 품사 정보를 추출할 때 조사와 어미의 문법 기능인 조사의 격 정보와 어미의 활용형 정보에 따라 몇 가지로 분류하고 정보를 추출하여 품사 태깅에 적용하면 조사와 어미를 분류하지 않은 품사열 만을 사용한 태깅 방법 보다 더 나은 성능을 얻을 수 있다.

• KIPS Transactions on Software and Data Engineering
• /
• v.5 no.11
• /
• pp.555-562
• /
• 2016

Semantic Role Labeling (SRL) is to determine the semantic relation of a predicate and its argu-ments in a sentence. But Korean semantic role labeling has faced on difficulty due to its different language structure compared to English, which makes it very hard to use appropriate approaches developed so far. That means that methods proposed so far could not show a satisfied perfor-mance, compared to English and Chinese. To complement these problems, we focus on suffix information analysis, such as josa (case suffix) and eomi (verbal ending) analysis. Korean lan-guage is one of the agglutinative languages, such as Japanese, which have well defined suffix structure in their words. The agglutinative languages could have free word order due to its de-veloped suffix structure. Also arguments with a single morpheme are then labeled with statistics. In addition, machine learning algorithms such as Support Vector Machine (SVM) and Condi-tional Random Fields (CRF) are used to model SRL problem on arguments that are not labeled at the suffix analysis phase. The proposed method is intended to reduce the range of argument instances to which machine learning approaches should be applied, resulting in uncertain and inaccurate role labeling. In experiments, we use 15,224 arguments and we are able to obtain approximately 83.24% f1-score, increased about 4.85% points compared to the state-of-the-art Korean SRL research.

• Annual Conference on Human and Language Technology
• /
• 1995.10a
• /
• pp.188-194
• /
• 1995

한글 문서에서는 일부 조사와 일부 어미가 자주 출현하며 그 외의 조사/어미는 출현 빈도가 낮을 것으로 추측되고 있다. 본 연구에서는 실험에 통해서 이러한 사실을 확인하고 자주 출현하는 통합형 조사와 어미의 빈도를 구하기 위하여 한국어 말뭉치에서 통합형 조사와 통합형 어미의 상대적 출현 빈도를 조사하였다. 통합형 조사의 상대적 출현 빈도를 조사한 결과 말뭉치의 분야에 따라 약간의 차이가 있으나 평균 상위 9개의 통합형 조사가 전체 조사의 70%를 차지하고 있으며 상위 20개, 32개, 69개의 통합형 조사가 각각 90%, 95%, 99%를 차지하고 있음을 확인하였다. 통합형 어말어미의 경우에는 상위 10개의 통합형 어말어미가 전체 어말어미의 70%를 차지하고 상위 33개, 54개, 117개의 통합형 어미가 각각 90%, 95%, 99%를 차지하고 있다. 본 논문에서는 조사, 어미의 상대적 출현 빈도에 따라 계층적으로 조사/어미 사진을 구성함으로써 형태소 분석 효율을 높이고 형태소 분석기가 다양한 응용 분야에 쉽게 적응할 수 있도록 하는 방법을 제안한다. 또한 통합형 조사, 어미의 상대적 출현 빈도는 미등록어 추정을 용이하게 하거나 형태론적 모호성을 해결할 때에도 유용하게 활용될 수 있음을 보인다.

• The Journal of Korean Association of Computer Education
• /
• v.3 no.2
• /
• pp.115-127
• /
• 2000

Korean error detectors of word processors have defects such as inconvenience that users choose one of error groups, lower detecting rate of 60%, and slow processing time. In this study, I proposed a resolution method of these defects. For these, I applied bidirectional longest match strategy for morphological analysis to improve processing time. I suggested dictionaries and several algorithms such as seperation of compound noun and assistant declinable words, correction of typing error to improve processing time and to guarantee correction accuracy. I also suggested a distinguishable method for dependent noun/suffix and Josa/Eomi where many ambiguities are generated, and a distinguishable method for Korean "로써/로서" to improve the reliability of the correction system.

• The Journal of the Korea Contents Association
• /
• v.17 no.4
• /
• pp.576-584
• /
• 2017

Unknown nouns which are not in a dictionary make problems not only morphological analysis but also almost all natural language processing area. This paper describes a recognition method for Korean unknown nouns using strings following nouns such as postposition, suffix and postposition, suffix and eomi, etc. We collect and sort words including nouns from documents and divide a word including unknown noun into two parts, candidate noun and string following the noun, by finding same prefix morphemes from more than two unknown words. We use information of strings following nouns extracted from Sejong corpus and decide unknown noun finally. We obtain 99.64% precision and 99.46% recall for unknown nouns occurred more than two forms in news of two portal sites.

• The Journal of Korean Association of Computer Education
• /
• v.6 no.2
• /
• pp.9-21
• /
• 2003

This paper proposes an automatic construction system for transliteration dictionary from English-Korean parallel corpus. The system works in 3 steps: it extracts all nouns from Korean documents as the first step, filters transliterated foreign word nouns out of them with the language identification method as the second step, and extracts the corresponding English words by using a probabilistic alignment method as the final step. Specially, the fact that there is a corresponding English word in most cases, is utilized to extract the purely transliterated part from a Koreans word phrase, which is usually used in combined forms with Korean endings(Eomi) or particles(Josa). Moreover, the direct phonetic comparison is done to the words in two different alphabet systems without converting them to the same alphabet system. The experiment showed that the performance was influenced by the first and the second preprocessing steps; the most efficient model among manually preprocessed ones showed 85.4% recall, 91.0% precision and the most efficient model among fully automated ones got 68.3% recall, 89.2% precision.

• Science of Emotion and Sensibility
• /
• v.13 no.3
• /
• pp.449-458
• /
• 2010

As occasion demands to obtain efficient information from many documents and reviews on the Internet in many kinds of fields, automatic classification of opinion or thought is required. These automatic classification is called sentiment classification, which can be divided into three steps, such as subjective expression classification to extract subjective sentences from documents, sentiment classification to classify whether the polarity of documents is positive or negative, and strength classification to classify whether the documents have weak polarity or strong polarity. The latest studies in Opinion Mining have used N-gram words, lexical phrase pattern, and syntactic phrase pattern, etc. They have not used single word as feature for classification. Especially, patterns have been used frequently as feature because they are more flexible than N-gram words and are also more deterministic than single word. Theses studies are mainly concerned with English, other studies using patterns for Korean are still at an early stage. Although Korean has a slight difference in the meaning between predicates by the change of endings, which is 'Eomi' in Korean, of declinable words, the earlier studies about Korean opinion classification removed endings from predicates only to extract stems. Finally, this study introduces the earlier studies and methods using pattern for English, uses extracted sentimental patterns from Korean documents, and classifies polarities of these documents. In this paper, it also analyses the influence of the change of endings on performances of opinion classification.