• Annual Conference on Human and Language Technology
• /
• 2000.10d
• /
• pp.110-115
• /
• 2000

자연 언어 처리의 구문 분석에서는 중의성 있는 결과가 많이 생성된다. 이러한 중의성을 해소하는데 어휘정보가 유용하다는 것은 잘 알려져 있으며, 이러한 어휘정보와 이를 이용한 중의성 해소에 관한 연구가 많이 이루어지고 있다. 본 논문은 한국어의 구문 구조 분석 시 부사에 의해 발생되는 중의성을 해소하기 위해 수식어 사전을 이용하여 구문 분석에서의 구조 중의성을 해소하였다. 수식어 사전의 어휘정보와 대상 말뭉치를 통해 각각의 부사에 대한 문법을 구성하고, 이를 이용하여 한국어 구문구조 분석에서 부사에 의해 발생되는 중의성을 줄일 수 있다.

• Korean Linguistics
• /
• v.65
• /
• pp.211-236
• /
• 2014

The aim of this study is to investigate categorical characteristics of 'mak' and their discourse functions through analyzed the prosodic features of 'mak'. The previous studies of 'mak' focused on grammatical or semantic characteristics, but this study focuses on the prosodic features of 'mak' based on speech data. As a result, adverb 'mak' and discourse marker 'mak' are distinguished from prosodic boundary, duration, pause and sort of number tonal patterns. Functions of discourse marker 'mak' is as follows: Maintenance of utterance, Attention, Delay, Expression negative manner. These functions have salient prosodic features related to their functions. Consequently prosodic features are important to analyze categorical characteristics and to establish functions of 'mak'.

• English Language & Literature Teaching
• /
• v.10 no.3
• /
• pp.41-66
• /
• 2004

Under the textbook-driven approach to English education in the Korean selling, the importance of the English textbook can not be overemphasized as the main source of learning materials. Recently, with the development of computer-based language corpora, the recognition of the importance of collocations and the availability of computerized databases of words have caused a resurgence and facilitation in the instruction of collocation. The primary purpose of the present study is to identify the characteristics of lexical collocation and the extent of its use in high school 10th-grade textbooks. From all the analyses, it is revealed that the language materials reflect various constructed collocation in the case of adjective+noun and noun+noun collocations in a natural context. However, verb+noun and adverb+verb collocations are not fully reflected. This is true for delexicalized verbs, and verb and adjective intensifiers. Also the language materials do not provide sufficient support for the lexical syllabus, even though all textbooks may be somewhat adequate in terms of vocabulary size. Finally, based on the analyses of the texts, the suggestions for English collocation instruction are made in the lexical approach.

• Annual Conference on Human and Language Technology
• /
• 2001.10d
• /
• pp.373-380
• /
• 2001

한국어에서 부사는 관형사와 구별하여 통사적으로 명사 등의 체언을 수식할 수 없다고 분석하는 것이 일반적이다. 의미적으로 상태성 명사와 수식관계를 가질 수 있는 정도부사의 경우에도 명사를 직접 수식하는 것이 아니라 그 명사를 보어로 취하는 지정사를 포함한 지정사구 전체를 수식한다는 것이 현재의 이론언어학에서의 입장이다. 본 논문에서는 말뭉치에 나타난 실제 문장을 기계적으로 처리하는 관점에서 정도부사의 수식을 받는 것은 지정사구가 아니라 상태성 명사 자체로 설정하고자 한다. 이러한 근거로서 말뭉치에서 추출한 실제 문장을 중심으로 정도부사의 수식을 받는 지정사구에 지정사가 생략되는 경우와 지정사구 형태가 아닌 다양한 명사구 형태 역시 정도부사의 수식을 받는 경우가 존재함을 보인다. 또한 정도 부사와 결합하는 명사들이 갖는 의미적 특성을 통해 정도부사와 명사와 결합시켜야 수식 관계의 처리에 용이함을 보이고 정도부사에 대한 이론적 설명에도 타당함을 보인다. 마지막으로 말뭉치에 나타난 정도부사의 수식을 받는 명사의 용례를 분석하여 빈도 및 하위 분류 특성을 살펴본다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2008.06c
• /
• pp.242-247
• /
• 2008

This paper discusses a design for Korean language parsing based on subcategorization. First, we discuss some important Korean grammar elements such as syntax category, josa, omi-conjugation, syntactic affix, dependent noun and also discuss subcategorization and expression patterns. Then, we show the basic structure of Korean language parsing process. The first stage scans the input sentence and processes article, noun phrase, numeral, josa, affix, dependent noun, adjective, omi-conjugation, adverb, auxiliary verb. The second stage deals with subcategorization patterns and expression patterns. The third stage processes the clauses and the fourth stage deals with SEA(Sentence Ending+Auxiliary).

• Annual Conference on Human and Language Technology
• /
• 1999.10e
• /
• pp.110-114
• /
• 1999

영한기계번역에서 영문의 복합문중 종속부사절이 한국어로 생성될 때, 종속절의 주절에 대한 의미에 따라 다양한 한국어 연결어미를 생성하게 된다. 주절의 의미를 보완하는 종속절은 그 연결어미에 의해 그 역할이 결정되는 것이다. 종속부사절을 이끄는 부사어는 연결어미로 재현되는데 기존의 사전을 기반으로 하는 기계번역시스템에서 사전에 있는 부사어의 표층어만을 이용하였기 때문에 그 생성결과가 만족스럽지 않았다. 영문의 부사어중의 일부는 한국어로 생성될 때 의미적 구분에 따라 여러 가지의 연결어미로 생성이 되어야 하는데 영어 해석에서 종속절의 의미 정보를 충분히 분석하지 못하는 경우가 많다. 본 연구에서는 종속부사절을 이끄는 영어의 부사어와 각 부사어가 생성될 때 필요한 한국어 연결어미를 정리하였다. 또한, 형태소와 구문 자질정보를 이용하여 여러 연결어미를 갖는 부사어의 경우에 하나의 연결어미를 선택한다.

• Asia Pacific Journal of Corpus Research
• /
• v.1 no.2
• /
• pp.27-39
• /
• 2020

This paper examines the frequency and distribution of the so-called "non-discourse marker functions" of so in essay writings produced by 200 L1 English speakers and 1,300 EFL learners in China, Japan, Korea, and Taiwan. Based on the data drawn from the International Corpus Network of Asian Learners of English, this study compares EFL learners and L1 English speakers' uses of so, identifying four grammatical uses, as (1) an adverb, (2) part of a fixed phrase, (3) a pro-form, and (4) a conjunction phrase specifying purpose. This study aims to show the wide variability among EFL learners with different L1s, identifying the tendency of usage both common among and specific to the sub-groups of EFL learners. The findings suggest that the learners demonstrate patterns distinctively different from those of L1 English speakers, indicating an underuse of so as a marker expressing "purpose" and an overuse as part of fixed phrases. Compared to L1 English speakers, the learners also tend to overuse so in the discourse marker functions, regardless of their L1s. The study proposes pedagogical implications focusing on discourse flow and diachronic aspects of so in order to understand its multifunctionality, although the latter is primarily suggested for advanced learners.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2009.01a
• /
• pp.773-777
• /
• 2009

In recent years, the field of synthesizing voice has been developed rapidly, and the technologies such as reading aloud an email or sound guidance of a car navigation system are used in various scenes of our life. The sound quality is monotonous like reading news. It is preferable for a text such as a novel to be read by the voice that expresses emotions wealthily. Therefore, we have been trying to develop a system reading aloud novels automatically that are expressed clear emotions comparatively such as juvenile literature. At first it is necessary to identify emotions expressed in a sentence in texts in order to make a computer read texts with an emotionally expressive voice. A method on the basis of the meaning interpretation that utilized artificial intelligence technology for a method to specify emotions of texts is thought, but it is very difficult with the current technology. Therefore, we propose a method to determine only emotion every sentence in a novel by a simpler way. This method determines the emotion of a sentence according to an emotion that words such as a verb in a Japanese verb sentence, and an adjective and an adverb in a adjective sentence, have. The emotional characteristics that these words have are prepared beforehand as a emotional words dictionary by us. The emotions used here are seven types: "joy," "sorrow," "anger," "surprise," "terror," "aversion" or "neutral."

• KIPS Transactions on Software and Data Engineering
• /
• v.9 no.2
• /
• pp.45-52
• /
• 2020

Since the opening of the national petition site, it has attracted much attention. In this paper, we perform topic analysis of the national petition site and propose a prediction model for answerable petitions based on deep learning. First, 1,500 petitions are collected, topics are extracted based on the petitions' contents. Main subjects are defined using K-means clustering algorithm, and detailed subjects are defined using topic modeling of petitions belonging to the main subjects. Also, long short-term memory (LSTM) is used for prediction of answerable petitions. Not only title and contents but also categories, length of text, and ratio of part of speech such as noun, adjective, adverb, verb are also used for the proposed model. Our experimental results show that the type 2 model using other features such as ratio of part of speech, length of text, and categories outperforms the type 1 model without other features.

• Journal of Digital Contents Society
• /
• v.19 no.1
• /
• pp.133-140
• /
• 2018

Accurate sentiment classification is an important research topic in sentiment analysis. This study suggests an efficient classification method of Korean sentiment using word2vec and ensemble methods which have been recently studied variously. For the 200,000 Korean movie review texts, we generate a POS-based BOW feature and a feature using word2vec, and integrated features of two feature representation. We used a single classifier of Logistic Regression, Decision Tree, Naive Bayes, and Support Vector Machine and an ensemble classifier of Adaptive Boost, Bagging, Gradient Boosting, and Random Forest for sentiment classification. As a result of this study, the integrated feature representation composed of BOW feature including adjective and adverb and word2vec feature showed the highest sentiment classification accuracy. Empirical results show that SVM, a single classifier, has the highest performance but ensemble classifiers show similar or slightly lower performance than the single classifier.