• Annual Conference on Human and Language Technology
• /
• 2000.10d
• /
• pp.269-276
• /
• 2000

본 연구에서는 사전의 뜻풀이말에서 의미 정보를 추출하고, 이 의미 정보를 확률 통계적 방법에 적용하여 의미 중의성을 해결하는 모델을 제안한다. 사전의 뜻풀이말에 동형이의어를 포함하고 있는 표제어와 뜻풀이말을 구성하는 보통 명사, 형용사와 동사를 의미 정보로 추출한다. 비교적 중의성이 자주 발생하는 9개의 동형이의어 명사를 대상으로 실험하였다. 학습에 이용된 데이터로 정확률을 실험하는 내부 실험의 결과, 체언류(보통 명사)와 용언류(동사, 형용사)의 가중치를 0.9/0.1로 주는 것이 가장 정확률이 높았다. 외부 실험은 국어 정보베이스와 ETRI 코퍼스를 이용하여 1,796문장을 실험하였는데, 평균 79.73%의 정확률을 보였다.

• Journal of KIISE:Software and Applications
• /
• v.28 no.9
• /
• pp.688-698
• /
• 2001

A homonym could be disambiguated by anther words in the context such as nouns, predicates used with the homonym. This paper proposes a homonym disambiguation system based on statistical semantic information which is extracted from definitions in dictionary. The semantic information consists of nouns and predicates that are used with the homonym in definitions. In order to extract accurate semantic information, definitions are used with the homonym in definitions. In order to extract accurate semantic information, definitions are classified into two types. One has hyponym-hypernym relation between title word and head word (homonym) in definition. The hyponym-hypernym relation is one level semantic hierarchy and can be extended to deeper levels in order to overcome the problem of data sparseness. The other is the case that the homonym is used in the middle of definition. The system considers nouns and predicates simultaneously to disambiguate the homonym. Nine homonyms are examined in order to determine the weight of nouns and predicates which affect accrutacy of homonym disambiguation. From experiments using training corpus(definitions in dictionary), the average accruracy of homonym disamguation is 96.11% when the weight is 0.9 and 0.1 for noun and verb respectively. And another experiment to meaure the generality of the homonym disambiguation system results in the 80.73% average accuracy to 1,796 untraining sentences from Korean Information Base I and ETRI corpus.

• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.6
• /
• pp.747-751
• /
• 2010

This paper proposes the method that increases the accuracy for tagging word meaning by creating sense tagged data automatically using machine readable dictionaries. The concept of meaning group is applied here, where the meaning group for each meaning of a target word consists of neighbor words of the target word. To enhance the tagging accuracy, the notion of concentration is used for the weight of each word in a meaning group. The tagging result in SENSEVAL-2 data shows that accuracy of the proposed method is better than that of existing ones.

• Journal of KIISE:Software and Applications
• /
• v.27 no.7
• /
• pp.769-783
• /
• 2000

In this paper, in order to disambiguate Korean noun word sense, we define a local context and explain how to extract it from a raw corpus. Following the intuition that two different nouns are likely to have similar meanings if they occur in the same local context, we use, as a clue, the word that occurs in the same local context where the target noun occurs. This method increases the usability of extracted knowledge and makes it possible to disambiguate the sense of infrequent words. And we can overcome the data sparseness problem by extending the verbs in a local context. The sense of a target noun is decided by the maximum similarity to the clues learned previously. The similarity between two words is computed by their concept distance in the sense hierarchy borrowed from WordNet. By reducing the multiplicity of clues gradually in the process of computing maximum similarity, we can speed up for next time calculation. When a target noun has more than two local contexts, we assign a weight according to the type of each local context to implement the differences according to the strength of semantic restriction of local contexts. As another knowledge source, we get a co-occurrence information from dictionary definitions and example sentences about the target noun. This is used to support local contexts and helps to select the most appropriate sense of the target noun. Through experiments using the proposed method, we discovered that the applicability of local contexts is very high and the co-occurrence information can supplement the local context for the precision. In spite of the high multiplicity of the target nouns used in our experiments, we can achieve higher performance (89.8%) than the supervised methods which use a sense-tagged corpus.

• KIPS Transactions on Software and Data Engineering
• /
• v.1 no.1
• /
• pp.31-42
• /
• 2012

In order to process the suffix derived nouns of Korean, most of Korean processing systems have been registering the suffix derived nouns in dictionary. However, this approach is limited because the suffix is very high productive. Therefore, it is necessary to analyze semantically the unregistered suffix derived nouns. In this paper, we propose a method to disambiguate homograph suffixes using Korean lexical semantic network(U-WIN) for the purpose of semantic analysis of the suffix derived nouns. 33,104 suffix derived nouns including the homograph suffixes in the morphological and semantic tagged Sejong Corpus were used for experiments. For the experiments first of all we semantically tagged the homograph suffixes and extracted root of the suffix derived nouns and mapped the root to nodes in the U-WIN. And we assigned the distance weight to the nodes in U-WIN that could combine with each homograph suffix and we used the distance weight for disambiguating the homograph suffixes. The experiments for 35 homograph suffixes occurred in the Sejong corpus among 49 homograph suffixes in a Korean dictionary result in 91.01% accuracy.

• Cross-Cultural Studies
• /
• v.52
• /
• pp.393-416
• /
• 2018

Negative Stripping Construction in English involves the disjunction but, the adverb not, and a constituent NP. This construction is an incomplete sentence although it delivers a complete sentential meaning. Interpretation of this construction may be ambiguous in that the constituent NP can either be construed as the subject, or as the complements including the object. To generate such sentences and resolve the issue of ambiguity, we propose a construction-based analysis under direct interpretation approach, rejecting previous analyses based on deletion approaches. In so doing, we suggest a negative stripping construction rule that can account for ambiguous meaning. This rule further can enable us to explain syntactic structures and readings of Negative Stripping Construction.

• Proceedings of the Korean Society for Cognitive Science Conference
• /
• 2000.06a
• /
• pp.383-389
• /
• 2000

본고는 한영번역기 개발을 위한 기초 작업으로 '와/과'구문에 나타나는 여러 가지 패턴을 정리하고 이들 구문에서 보이는 중의성 문제를 해결하고자 하였다. 이러한 작업을 위해서는 자료 수집 및 분석이 우선이기 때문에 코퍼스에서 '와/과'구문을 뽑아서 분석하여 규칙을 마련하였다. 여기에서 사용된 자질연산문법(FCG)은 자연언어처리를 위한 문법으로 변형규칙과 수형도의 개념 없이 자질을 이용한 연산 체계로서 언어처리를 하고자 하는 문법이다. 이 이론을 바탕으로 규칙을 세우고 실제 언어 자료를 뽑아서 테스트를 하여 95%의 성공률을 보여주었다. 그러나 여기서의 연구는 '와/과'구문의 처리를 위한 가장 뼈대가 되는 기초연구이며, 앞으로 좀 더 많은 처리가 이루어져야 하리라 생각된다.

• Annual Conference on Human and Language Technology
• /
• 1998.10c
• /
• pp.269-275
• /
• 1998

구문 분석에서의 중의성을 해결하는 데 어휘정보가 유용하다는 것은 잘 알려져 있다. 그러나 기존의 어휘정보 구축 방법들은 많은 수작업을 요구하거나, 자동으로 구축하는 경우에는 어휘 자체를 그대로 사용함에 따라 심각한 자료 회귀성 문제가 발생했다. 본 논문에서는 구문 분석에서의 중의성 해소를 위해 원시 코퍼스와 시소러스를 이용하여 개념 수준(conceptual-level)의 일반화된 술어-인자 어휘정보를 자동으로 구축하고, 이를 파서에 적용하는 방법을 제안하고자 한다. 제안한 방법으로 구축한 일반화된 어휘정보를 파서를 이용하여 명사구의 지배소 결정 실험에 적용하여 본 결과, 정확도가 85.9%에서 91.5%로 향상되었다. 또, 미지격 결정 실험에 대해서는 86.32 %의 격 결정 성공률을 보여주었다.

• Journal of KIISE:Software and Applications
• /
• v.33 no.12
• /
• pp.1073-1089
• /
• 2006

The goal of Natural Language Processing(NLP) is to make a computer understand a natural language and to deliver the meanings of natural language to humans. Word sense Disambiguation(WSD is a very important technology to achieve the goal of NLP. In this paper, we describe a technology for automatic homonyms disambiguation using both Mutual Information(MI) and a Sense-Tagged Compound Noun Dictionary. Previous research work using word definitions in dictionary suffered from the problem of data sparseness because of the use of exact word matching. Our work overcomes this problem by using MI which is an association measure between words. To reflect language features, the rate of word-pairs with MI values, sense frequency and site of word definitions are used as weights in our system. We constructed a Sense-Tagged Compound Noun Dictionary for high frequency compound nouns and used it to resolve homonym sense disambiguation. Experimental data for testing and evaluating our system is constructed from QA(Question Answering) test data which consisted of about 200 query sentences and answer paragraphs. We performed 4 types of experiments. In case of being used only MI, the result of experiment showed a precision of 65.06%. When we used the weighted values, we achieved a precision of 85.35% and when we used the Sense-Tagged Compound Noun Dictionary, we achieved a precision of 88.82%, respectively.

• Annual Conference on Human and Language Technology
• /
• 2002.10e
• /
• pp.171-175
• /
• 2002

접사 처리는 복합명사 분석에서 중요한 문제인데 접사가 복합명사에 포함되어 있을 경우 여러 중의적 형태로의 분석이 가능하고 또한 미등록어 문제를 발생시킬 수 있기 때문이다. 단순한 접사 사전 정보만으로는 효율적인 분석을 수행할 수 없으므로 추가적인 정보가 필요하다. 본 논문에서는 접사로 인한 복합명사의 분석 중의성을 해소하기 위하여 명사의 의미 정보를 이용하는 방법에 대해 제안한다. 명사 의미 정보는 시소러스의 의미계층 정보로 최상위 계층 정보와 하위 4계층의 정보로 구성된다. 명사+접미사 형태의 의미 결합 정보를 구한 추, 접미사를 포함하는 복합명사의 단위 명사들 간의 의미 결합 정보를 구한다. 이렇게 구해진 명사들 간의 의미 결합 정보는 사전 정보에 추가되며 접사로 인한 중의적 분석 문제가 발생할 경우 명사들 간의 결합 정보를 이용하여 올바른 분석 후보를 선택한다.