계열관계에 기반한 단어 분석과 단어 형성

채현식(Chae, Hyun-Sik)(蔡炫植) 형태론 2012 형태론 Vol.14 No.2

단어는 통합관계와 계열관계에 바탕을 두고 구조화된다. 단어의 분석과 형성 과정은 통합관계에 기반하여 접근할 수도 있고 계열관계에 기반하여 접근할 수도 있다. 통합관계에 기반한 접근법으로 생성형태론의 단어형성규칙을 들 수 있다. 그러나 단어형성규칙으로는 단어의 분석과 형성 과정을 적절히 설명하기 어렵다. 단어를 분석하고 형성하는 과정은 단어들의 계열관계에 기반할 때 적절히 설명된다. 단어의 분석은 계열관계에 있는 단어들 사이의 비교를 통해서 이루어지며 단어의 형성은 기존 단어, 또는 기존 단어들의 활성화로 창발된 단어틀을 본보기로 한 유추의 방법으로 이루어진다. 한편 최근 규칙을 화자의 단어형성능력의 형식화로 규정하고 단어 형성에 관련해서 기존의 유추와 규칙의 이분법적 대립을 지양하려는 논의가 있었다. 심리규칙으로서의 단어형성규칙은 생산성, 제약, 의미론적 조건 등과 무관하게 설정될 수 있다는 것이다. 인지 과정으로서의 규칙과 유추의 본질에 대한 진지한 논의가 지속될 필요가 있다. This paper aims to reveal the mechanism of human competence to analyze the internal structures of words and to form new words. Words are structured by two types of relationships: syntagmatic and paradigmatic relationships. The former is related to linear concatenations of elements of complex words and the latter is related to various connections within words in the mental lexicon. The competence to analyze complex words into their constituent elements and to form new words may be explained by an approach based on the syntagmatic relationships between elements of words or by an approach based on the paradigmatic relationships within words. The representative approach based on the syntagmatic relationships between words is a generative morphology in which word-formation rules are a powerful mechanism with which to analyze existing words and to form new words. However, it should be noted that word-formation rules are inadequate devices for such a purpose. In contrast, the approach based on the paradigmatic relationships within words can offer a reliable solution for problems in word-formation rules. In the approach based on the paradigmatic relationships within words, the internal structures of complex words are analyzed into their constituent elements by comparing connected words in which common elements are connected to one another. New words are then formed based on existing individual words, group of words, or word-schema concepts, which has emerged from paradigmatically related words.

단어 기억의 신경과학적 기제

채현식 ( Hyun Sik Chae ) 전북대학교 인문학연구소 2016 건지인문학 Vol.15 No.-

The aim of this thesis is to look out upon the representations of our knowledge about words. In particular we focus on how is our knowledge of words encoded and stored in the brain on the basis of neuroscientic surveys. The issues of word memory representation and the answers about those issues in neuroscience are as follows. First, what is the neural mechanism for word memories? It does not merely mean electro-chemical changes in synapses to form word memories. Instead, forming word memories are followed by changes in the level of neural cell such as gene expression and protein synthesis. The neurobiological mechanisms which explain these changes are synaptic plasticity, Hebb`s rule, long term protentiation and consolidation. Second, What are the parts of brain which is involved in word memory? The major parts of brain which involved in word memory are the cerebral cortex and hippocampal structure. Third, what is the process of encoding new words in the brain? New words are distributedly encoded in cerebral cortex. But before the new words are fixed in the cortex, hippocampal structure intermediate words in short-term memory and long-term memory. Fourth, where are the areas that the informations of word are represented, and how are the informations of word represented in the brain? The perceptual and motor-related informations of word are represented in sensory and motor areas of brain. The abstract informations of word are represented in the association areas which are the parts of sensory-motor areas. It is not uncertain whether there are specialised brain areas for taxonomic categories of words.

어휘부란 무엇인가-인접 학문 분야와와 관계를 중심으로

채현식 ( Chae Hyun-sik ) 국어학회 2013 국어학 Vol.66 No.-

이 글은 언어학과 관련을 맺는 인접 학문 분야와 어휘부 사이의 관계를 살펴봄으로써 어휘부가 지닌 성격을 조망해 보는 데 목적이 있다. 심리언어학의 관점에서 단어 저장소로서의 어휘부의 성격은 기억과 관련해서 살펴볼 수 있다. 어휘항목으로서의 단어에 대한 지식은 의미기억에 대응된다. 음운 규칙과 통사 규칙은 절차기억에 대응된다. 만약 단어형성규칙을 상정한다면, 규칙 안에 접사나 어휘적 요소가 포함된 것은 의미기억과 절차기억의 성격을 함께 갖는다. 그러나 변항만으로 이루어진 추상적인 합성어형성규칙은 기억의 관점에서 규칙으로서의 지위를 인정받기 어렵다. 규칙이 아니라 틀을 상정한다면 틀은 의미기억에 대응한다. 한편 뇌신경과학의 관점에서 단어의 정보는 대뇌 피질의 여러 영역에 분산 표상되어 있고 병렬적으로 처리된다. 특히 단어의 음운 정보와 개념적 의미 정보는 측두엽에 기억된다. 의미지식의 기억 과정에는 해마가 관여하는 반면 통사 규칙과 같은 절차지식의 기억 과정에는 기저핵이 관여한다. 심리언어학과 뇌신경과학의 관점에서 어휘부와 통사부는 독립된 모듈로 존재한다. 어휘부에서는 단어들의 연결을 통해 단어 층위와는 다른 새로운 질서가 만들어진다. 어휘부의 그러한 특징은 복잡계 이론의 ‘자기조직화’와 ‘창발’의 개념으로 설명할 수 있다. 또한 어휘부는 환경에 적응하며 그 과정에서 비선형적인 특징을 보이기도 하는데 이 역시 어휘부가 복잡계임을 뜻한다. This aim of this paper is to look out upon the properties of mental lexicon by observing the relations among linguistics and psycholinguistics and neuroscience. In psycholinguistics the properties of lexicon can be observed in the respect of memory. The knowledge of words as lexical items corresponds semantic memory, whereas the phonological and syntactic rules correspond procedural memory. If word-formation rules could be assumed, the rules with affixes or lexical item would have the both side of characteristics of semantic and procedural memories. But the existence of rule will be denied if we assume the compounding rule that only have variables without lexical items. On the other hand we can assume word schema instead of rules, then the word schema corresponds semantic memory. In neuroscience the information of words is distributed in various regions on brain cortex and is parallelly processed. The information of phonological and conceptual semantic information of words is memorized in temporal lobe. Hippocampus is concerned in the memory of semantic information whereas basalganglia is concerned in the memory of procedural knowledge. Pshcholinguistics and neuroscience offer the evidences that lexicon and syntax are independent modules. A new orders different from word-level are formed in the lexicon by inter-connections among words. Such properties of lexicon can be explained by introducing the conceptions of ‘emergence’ and ‘self-organization’. The lexicon is always adaptive to circumstances and sometimes shows non-linear properties. It is also the characteristic of lexicon as a complex system.

탄소봉 실험을 이용한 조립토 Piled Raft 기초의 거동 연구

채현식(Chae Hyun Sik),김홍택(Kim Hong Taek),박사원(Park Sa Won),신진택(Shin Jin Taek),신방웅(Shin Bang Woong) 대한토목학회 2000 대한토목학회 학술발표회 논문집 Vol.2000 No.2

합성명사에서의 의미 전이와 관습화

채현식(Chae Hyunsik) 한국언어문학회 2006 한국언어문학 Vol.58 No.-

  The aim of this paper is to investigate when metaphorical meaning occurs in compound nouns. In compound words, transferred meaning may occur during or before the process of compounding. We treat this matter in the term of psychological processes which operate in production and interpretation of compounding words. We argue ‘conventionalization’ of the transferred meaning offers the key of solution. In metaphoric expressions, if the transferred meanings have been conventionalized to the lexical meaning of source nouns, meaning-transfer occurs before compounding. In the rest metaphoric expressions, meaning-transfer occurs during compounding. Conventionalizations of the transferred meaning take place in two domains, one of which is individual word level, and the other is schema level. The consistency of transferred meanings and the frequencies of source nouns play a critical roles in conventionalization. In conventionalization of transferred meanings of individual word level, token frequency of those words is important factor. The conventionalization of transferred meanings in schema level depends on type frequency of members of the schema. (Kunsan University)

명사의 형태론

채현식(Hyun Sik Chae) 한국어학회 2014 한국어학 Vol.62 No.-

The purpose of this paper is to describe some issues of morphological phenomena of noun and to present my opinions for those issues. First, we deal with two issues concerning category of noun. One of them is to establish the part of speech of so-called ``adnominal-noun``, and the other of them is to distinguish noun compound from noun phrase. Second, the issue of the place of word-formation of noun compound is dealt, and I argue noun compounds are formed in both lexicon and syntax. Third, we discuss the head of noun. We argue that the head of noun are to be classified semantic head and categorial head in order to explain morphological phenomena of noun.

용례 기반 이론에서의 어휘 지식 표상

채현식(Chae, Hyunsik) 도서출판 박이정 2009 형태론 Vol.11 No.2

화자는 복합어의 내부구조를 분석하여 구성요소를 인식할 수 있으며 복합어들 사이에 존재하는 형태론적 규칙성을 알고 있다. 화자의 이러한 어휘지식은 비교, 추상화, 네트워크 형성이라는 일반 인지 능력에 기초하여 만들어진 형태론적 틀로 표상된다. 의존적인 구성요소인 접사나 어근은 독자적으로 표상되지 못하며 이들을 공유한 단어 계열체로부터 추상된 틀의 일부로 표상된다. 복합어들 사이에 존재하는 형태론적 규칙성과 복합어의 내부구조는 용례들로부터 추상된 틀의 형식으로 표상된다. 형태론적 규칙성을 결정하는 요소는 틀의 강도이다. 복합어들의 유형빈도가 높을수록 틀의 강도가 커진다. 복합어의 내부구조를 분석할 때, 틀의 강도와 함께 틀의 상세성이 중요한 분석 기준이 된다. 틀의 표상에 복합어와 공유하는 언어 요소가 많이 포함될수록, 의미정보가 구체적으로 명세될수록 틀의 상세성이 높아진다. 하나의 복합어에 대해 서로 다른 내부구조를 반영하는 두 개의 틀이 만들어질 수 있다. 두 개의 틀 중 강도가 크고 상세성이 높은 틀이 경쟁에서 이기게 된다. 화자는 경쟁에서 승리한 틀에 기대어 복합어의 내부구조를 분석한다. Speakers can analyze the internal structures and the constituent elements of complex words. Speakers also can recognize the morphological regularities that exist in complex words. According to the Usage-based theory, lexical knowledge of speakers is represented by morphological schemata based on comparison, abstraction, and networking words, which are parts of cognitive abilities of humans. The constituent elements of a complex word are recognized by comparing that word with other complex words which contain them in common. Affixes and roots do not have their own representations. Rather, they are represented as parts of schemata abstracted from complex words. Morphological regularities and internal structures of complex words are represented by schemata abstracted from words in use. Morphological regularities are determined by the strength of a schema. The strength of a schema increases as the type frequencies of members of the schema increase. Along with the strength of a schema, the specificity of a schema is decisive in analyzing internal structures of complex words. The more concrete linguistic elements and detailed meaning information a schema has, the more these things become schema specific. A word can be analyzed as having different internal structures with more than one schema. The schema which gets a higher score in strength and specificity wins in competition, and speakers analyze an internal structure of the word based on that schema.

정보의 처리와 표상의 측면에서 본 괄호매김역설

채현식(Chae Hyunsik) 한국언어문학회 2010 한국언어문학 Vol.74 No.-

The hierarchical structure of Complex words has been analyzed by the tool called "Immediate Constituent(IC)". Sometimes IC analysis of Complex words lead to asymmetry between morphological structure and meaning structure of Complex Words, which is known to us bracketing-paradox. The aim of this thesis is to make clear that bracketing-paradox does not exist from viewpoint of Cognitive Grammar, and the examples of bracketing -paradox can not be counter-evidence against symmetry between morphological structure and meaning structure of complex words. The bracketing-paradox is logical conclusion of serial processing and local representation of information, which is the basic stance of Generative morphology. Depending on serial processing, complex words are analyzed to only one structure, and depending on local representation of information, the argument structures of verbs are only represented in verb, not in other constituents of complex words. Two structures could not be compatible, and the bracketing-paradox is inevitable. However bracketing-paradox does not occur on the assumption of Cognitive Grammar, because it suppose parallel processing and distributed representation of information. Parallel processing of information permits over two analysis of complex words, and distributed representation of information permits sharing of argument structure with verbs and other constituents of complex words. Hence bracketing paradox does not occur in the theoretical system of Cognitive Grammar.

Multi-Domain 과 행렬 보간법을 이용한 강 하구와 연안의 2 차원 수치해석

채현식(H. S. Chae) 한국전산유체공학회 1997 한국전산유체공학회지 Vol.2 No.1

This paper presents a two-dimensional horizontal implicit model to general circulation in estuaries and coastal seas. The model is developed in non-orthogonal curvilinear coordinates system, using the Interpolatiug Matrix Method (IMM), in combination with a technique of multi-domain. In the propose model, the Saint-Venant equations are solved by a splitting-up technique, in the successive steps; convection, diffusion and wave propagation. The ability of the proposed model to deal with full scale nature is illustrated by the interpretation of a dye-tracing experiment in the Gironde estuary.

배달 앱 온·오프라인 서비스 품질과 전환비용, 고객 만족, 그리고 지속적 사용의도에 관한 실증적 연구

채현식(Hyun-Sik Chae),김은정(Eun-Jung Kim),김종원(Jong-Weon Kim) 한국산업정보학회 2021 한국산업정보학회논문지 Vol.26 No.6

본 연구는 4차 산업혁명시대와 Covid 19로 인해 비대면 라이프스타일이 가속화됨에 따라 배달 앱을 이용하는 사용자의 행태를 분석하였다. 구체적으로 본 연구는 배달 앱 서비스를 이용하는 고객의 지각된 서비스 품질이 전환비용, 고객 만족과 지속적 사용의도에 미치는 영향을 살펴보고, 그 결과를 토대로 배달 앱을 이용하는 이용자(고객), 배달 앱에 가입하여 운영하는 입점자(음식점), 그리고 고객과 음식점을 연결하여 주는 플랫폼(배달의 민족 등)에 유용한 정보와 시사점을 제공하는 것이 주요 목적이다. This study investigates the behavior of users using delivery apps as non-face-to-face lifestyles accelerated due to the 4th industrial revolution era and Covid 19. Specifically, this study examines the effect of customers" perceived service quality on the conversion cost, customer satisfaction, and continuous use intention. Based on the results, the main purpose of this study is to provide useful information and implications for entrants (restaurants) who have signed up and operated and platforms that connect customers, restaurants, and delivery people.