• Annual Conference on Human and Language Technology
• /
• 1999.10e
• /
• pp.265-271
• /
• 1999

이 논문은 한국어 관계관형절(relative adnominal clause)의 전산처리에 적합한 통사 의미 표상 모형을 제시하고, 그 결과를 전산적 구현을 통해서 검증하는 것이 목적이다. 이를 위해 이 논문에서는 다음의 두 가지 문제를 중심으로 관계관형절의 통사 의미 표상과 전산적 구현 문제를 다룬다. 첫째, 관계관형절의 수식을 받는 머리 명사(head noun)는 관계관형절과 모문(matrix sentence)에서 각각 다른 의미역할을 하는 논항이다. 즉, 하나의 논항이 두 개의 의미역을 표상한다. 이 논문의 첫째 과제는 이러한 관계관형절 구문에서 머리 명사의 이중의미역을 표상하는 방법을 모색하는 것이다. 둘째, 관계관형절이 일항술어로 구성될 때, 서술어 단독으로 머리 명사를 수식할 수 있을 뿐만 아니라, 주격중출 구문을 관계화하여 미리 명사를 수식할 수도 있다. 그러나 모든 일항술어가 주격중출 구문을 구성할 수 있는 것은 아니기 때문에 주격중출 구문의 관계화가 가능한 경우와 그렇지 않은 경우를 구별할 필요가 있다. 이 논문의 둘째 과제는 이러한 주격중출 구문의 관계화와 그 표상의 문제를 다루는 것이다. 이 논문에서는 이러한 문제들을 단순히 기술하는 데 그치지 않고 전산 구현을 통해 문제해결을 제시한다. 이를 위해 구현 도구로 C-언어를 보강하여 개발한 문법개발 도구언어인 말라가(Malaga)를 사용하며, 분석결과를 자질구조(feature structure)로 명시하여 그 타당성을 검토한다.

• The KIPS Transactions:PartB
• /
• v.10B no.2
• /
• pp.219-228
• /
• 2003

This paper presents an efficient construction method of determination rules of thematic roles from syntactic relations in Korean language processing. This process is one of the main core of semantic analysis and an important issue to be solved in natural language processing. It is problematic to describe rules for determining thematic roles by only using general linguistic knowledge and experience, since the final result may be different according to the subjective views of researchers, and it is impossible to construct rules to cover all cases. However, our method is objective and efficient by considering large corpora, which contain practical osages of Korean language, and case frames in the Sejong Electronic Lexicon of Korean, which is being developed by dozens of Korean linguistic researchers. To determine thematic roles more correctly, our system uses syntactic relations, semantic classes, morpheme information, position of double subject. Especially by using semantic classes, we can increase the applicability of the rules.

• Journal of KIISE:Software and Applications
• /
• v.30 no.5_6
• /
• pp.466-475
• /
• 2003

To build an efficient Question-Answering (QA) system, a question type classifier is needed. It can classify user's queries into predefined categories regardless of the surface form of a question. In this paper, we propose a question type classifier using a Support Vector Machine (SVM). The question type classifier first extracts features like lexical forms, part of speech and semantic markers from a user's question. The system uses $X^2$ statistic to select important features. Selected features are represented as a vector. Finally, a SVM categorizes questions into predefined categories according to the extracted features. In the experiment, the proposed system accomplished 86.4% accuracy The system precisely classifies question type without using any rules like lexico-syntactic patterns. Therefore, the system is robust and easily portable to other domains.

• Journal of KIISE:Software and Applications
• /
• v.31 no.12
• /
• pp.1649-1659
• /
• 2004

Text Generation is a process of generating comprehensible texts in human languages from some underlying non-linguistic representation of information. Among several sub-processes for text generation to generate coherent texts, this paper concerns referring expression generation which produces different types of expressions to refer to previously-mentioned things in a discourse. Specifically, we focus on pronominalization by zero pronouns which frequently occur in Korean. To build a generation model of referring expressions for Korean, several features are identified based on grammatical information and cost-based centering model, which are applied to various machine learning techniques. We demonstrate that our proposed features are well defined to explain pronominalization, especially pronominalization by zero pronouns in Korean, through 95 texts from three genres - Descriptive texts, News, and Short Aesop's Fables. We also show that our model significantly outperforms previous ones with a 99.9% confidence level by a T-test.

• Journal of KIISE:Software and Applications
• /
• v.30 no.11
• /
• pp.1062-1071
• /
• 2003

A genre-based classification means classifying documents by the purpose for which they were written, not by the semantics or subject areas. Most genre classifying methods in the past were based on the existing documents categorization algorithms and ineffective for feature selections, resulting in low quality classification results. In this research, we propose a new method for automatic classification of digital documents by genre. The genre classifier we developed uses the deviation statistic between the genre-revealing term frequencies and between the subject-revealing term frequencies within a genre. We collected Web documents to evaluate the proposed genre classification method. The experimental results show that the proposed method outperforms a direct application of a kai-square feature selection and bayesian classifier often used for subject classification by proving an excellent accuracy of about 30 percent.

• Journal of the Korea Convergence Society
• /
• v.9 no.12
• /
• pp.47-52
• /
• 2018

Named Entity Recognition is a system that extracts entity names such as Persons(PS), Locations(LC), and Organizations(OG) that can have a unique meaning from a document and determines the categories of extracted entity names. Recently, Bi-LSTM-CRF, which is a combination of CRF using the transition probability between output data from LSTM-based Bi-LSTM model considering forward and backward directions of input data, showed excellent performance in the study of object name recognition using deep-learning, and it has a good performance on the efficient embedding vector creation by character and word unit and the model using CNN and LSTM. In this research, we describe the Bi-LSTM-CNN-CRF model that enhances the features of the Korean named entity recognition system and propose a method for constructing the traditional culture corpus. We also present the results of learning the constructed corpus with the feature augmentation model for the recognition of Korean object names.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.23 no.6
• /
• pp.1103-1110
• /
• 2013

This paper proposes a method of identifying ownership of personal information in Social Network Service. In detail, the proposed method automatically decides whether any location information mentioned in twitter indicates the publisher's residence area. Identifying ownership of personal information is necessary part of evaluating risk of opened personal information online. The proposed method uses a set of decision rules that considers 13 features that are lexicographic and syntactic characteristics of the tweet sentences. In an experiment using real twitter data, the proposed method shows better performance (f1-score: 0.876) than the conventional document classification models such as naive bayesian that uses n-gram as a feature set.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2005.11a
• /
• pp.669-672
• /
• 2005

단어가 지니는 의미를 이미지로 처리하면 그 단어의 이미지가 단순화되면서 높은 인식률을 가질 수 있다는 장점을 지닌다. 이러한 장점을 이용하여 언어장애인들을 위한 유용한 보완대체 의사소통 학습도구에 하나로 이미지 전자 사전을 구축하고자 한다. 우선 동사와 조합되는 용어들의 패턴들을 면밀히 조사하여 그 패턴들을 영역과 자질의 카테고리로 분류하고, 그 카테고리에 속하는 기본 데이터들을 정리하여 분류된 데이터를 하위범주화 방식을 통해 검색을 보다 용이하게 하였다. 더욱이 언어장애인들이 많이 쓰이는 단어를 조사하고, 그 단어를 중심으로 한 모듈을 이용하여, 각각에게 해당되는 이미지를 수집함으로 단어들의 의미를 표현하고 인식할 수 있도록 하는 인터페이스를 구축하는데 중점을 두었다. 또한 언어장애인이 직접 명사와 동사를 조합하여 그 완성여부를 검토할 수 있는 학습기능을 추가함으로 인해 보다 실생활에 유용하고 교육적인 이미지 전자 사전을 구축하였다.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.11 no.8
• /
• pp.3021-3027
• /
• 2010

Example-based chatBot generates a response to user's utterance by searching the most similar utterance in a collection of dialogue examples. Though finding an appropriate example is very important as it is closely related to a response quality, few studies have reported regarding what features should be considered and how to use the features for similar utterance searching. In this paper, we propose a machine learning framework which uses various linguistic features. Experimental results show that simultaneously using both semantic features and lexical features significantly improves the performance, compared to conventional approaches, in terms of 1) the utilization of example database, 2) precision of example matching, and 3) the quality of responses.

• Annual Conference on Human and Language Technology
• /
• 1998.10c
• /
• pp.222-229
• /
• 1998

본 논문에서는 X-바 이론을 한국어에 적용하여 서로 다른 범주들간의 구조적 일반성을 파악하고, 한국어에 가능한 규칙만을 허용하여 불가능한 규칙을 배제시킬 수 있는 틀을 제시하고자 한다. 한국어가 비중심어간 어순이 자유롭고 기능어가 발달했다는 점을 고려하여, 중심어와 보충어 관계 중심의 기존 X-바 이론을 통사적 파생과 의미적 파생, 수식 및 하위범주의 관계를 중심으로 변형한다. 또한, 한국어의 빈번한 생략현상과 부분 자유 어순에 효과적으로 대응할 수 있도록 이진결합 중심의 CNF(Chomsky Normal Form)를 따른다. 제안하는 자질기반의 한국어 구구조 문법은 직관적이고 간단하면서도 대부분의 문장을 처리할 수 있을 만큼 표현력이 뛰어나다는 장점이 있다. 신문기사에서 454문장을 추출하여 실험한 결과, 약 97%의 문장에 대해 올바른 구문 분석 결과를 생성할 수 있음을 보였다.