간행물

서강대학교 언어정보연구소> 언어와 정보사회

언어와 정보사회 update

Language & information society

  • : 서강대학교 언어정보연구소
  • : 어문학분야  >  언어학
  • : KCI등재
  • :
  • : 연속간행물
  • : 연3회
  • : 1598-1886
  • : 2713-6817
  • :

수록정보
수록범위 : 1권0호(2000)~45권0호(2022) |수록논문 수 : 441
언어와 정보사회
45권0호(2022년 03월) 수록논문
최근 권호 논문
| | | |

KCI등재

1서울말 어휘적 장단의 비음운화 재고: 선행 연구에 대한 비판적 검토를 통하여

저자 : 김지은 ( Kim Jieun )

발행기관 : 서강대학교 언어정보연구소 간행물 : 언어와 정보사회 45권 0호 발행 연도 : 2022 페이지 : pp. 1-27 (27 pages)

다운로드

(기관인증 필요)

초록보기

Although the literature shows that vowel length distinction in the Seoul dialect has already disappeared, some recent studies claim that it still exists. The seemingly contradictory conclusions in previous studies lead to the question, “Do speakers happen to acquire vowel length distinction as they get aged?” (Jeong 2000). In this study, I examine 1) when Seoul Korean started to lose vowel length distinction, 2) whether/how Seoul Korean still retains the distinction, 3) and why previous studies drew different conclusions. In sum, I argue that 1) Seoul Korean began to lose vowel length distinction at least at the beginning of the 20th century, 2) vowel length distinction still for some speakers and in some lexicons, and 3) the previous studies drew different conclusions depending on the viewpoint and methodology.

KCI등재

2우즈베크인 한국어 학습자의 단모음 발음 양상과 특성

저자 : 하영우 ( Ha Young-woo ) , 안성민 ( An Sung-min )

발행기관 : 서강대학교 언어정보연구소 간행물 : 언어와 정보사회 45권 0호 발행 연도 : 2022 페이지 : pp. 29-59 (31 pages)

다운로드

(기관인증 필요)

초록보기

The purpose of this study is to examine the patterns and characteristics of monophthongs by comparing the pronunciation of Korean and Uzbek vowels by Uzbek learners of Korean and to provide basic data for instruction. For this purpose, the patterns and characteristics of monophthong pronunciation of Uzbek learners are revealed through a contrastive analysis of Uzbek and Korean. The pronunciation of monophthongs in Korean by Uzbek learners shows that the vowel space is generally expanded in word utterances, and the tendency of space overlap between vowels is strong. First, it is confirmed that the spatial overlapping pattern of the front vowels /E/ and /ㅣ/ is directly influenced by the mother tongue and /E/ is articulated close to /ㅣ/. The space for the vowel /ㅡ/ expands to meet the space of /ㅏ/, reaching a low vowel. The vowel spatial overlap of /ㅓ/ appears partially with /ㅏ/ in the case of female speakers. On the other hand, in the case of male speakers, /ㅓ/ has a wide overlap with /ㅗ/, and there is also some overlap of /ㅜ/ and /ㅡ/. Female speakers show overlap of /ㅗ/ and /ㅜ/ most of the time, whereas male speakers show overlap of /ㅗ/ to a greater extent with /ㅓ/ than with /ㅜ/. The overlap appeared to be much more widespread.

KCI등재

3준구어 원시균형 말뭉치의 장르별 구어성

저자 : 이규호 ( I Gyuho )

발행기관 : 서강대학교 언어정보연구소 간행물 : 언어와 정보사회 45권 0호 발행 연도 : 2022 페이지 : pp. 61-81 (21 pages)

다운로드

(기관인증 필요)

초록보기

Although quasi-spoken language has the weakness of being an artificial language, not a pure spoken language, it reveals a colloquial character that is distinct from written language. The colloquiality of quasi-spoken language is found at all levels of grammar in phonology, morphology, syntax, and discourse. How does the degree of colloquiality vary with respect to the genres of drama, movie, play, and musical?
In all genres, the abbreviated forms were preferred over the basic forms. In the drama/movie genre, the connective ending of the final function, the short negative sentence, the pronoun 'ni'(colloquial you), and the discourse marker 'sure, but' were preferred. In the play/musical genre, informal ending, the long negative sentence, the pronouns 'ne'(you), and discourse marker 'kuriko(and), ani(no)' were preferred.

KCI등재

4중국어 동사 중첩형의 상 기능과 문법화 과정

저자 : 백은희 ( Paek Eun-hee )

발행기관 : 서강대학교 언어정보연구소 간행물 : 언어와 정보사회 45권 0호 발행 연도 : 2022 페이지 : pp. 83-107 (25 pages)

다운로드

(기관인증 필요)

초록보기

'V yi V' and 'VV' is verb reduplication forms in Chinese, which have the functions of the delimitative aspect and the habitual aspect. The delimitative aspect means doing something to a small extent or for a short period of time; the habitual aspect means actions take place habitually or repeatedly. The reason why the same linguistic form of the verb reduplication has these two quite different functions can be explained by their grammaticalization paths. 'V+numeral+V' appeared in the Five Dynasties(五代), in which the numeral represented the number of repetitions. 'V yi V' appeared in the South Song Dynasty(南宋) and 'VV' appeared in the Ming Dynasty(明). 'V yi V' is originated from 'V+numeral+V', but the numeral meaning of yi has weakened; 'V V' is the form of phonological erosion of 'V yi V'. 'V yi V' and 'VV' have the functions of delimitative aspect, but habitual aspect is represented only by 'VV', which means 'VV' is the latest one in the grammaticalization path.

KCI등재

5평양 지역어의 음운 체계와 음운 현상

저자 : 소신애 ( So Shin-ae )

발행기관 : 서강대학교 언어정보연구소 간행물 : 언어와 정보사회 45권 0호 발행 연도 : 2022 페이지 : pp. 109-144 (36 pages)

다운로드

(기관인증 필요)

초록보기

This paper discusses the vowel system, consonant system, and related phonological phenomena of the Phyengyang subdialect after the mid-20th century. The properties of the Phyengyang subdialect identified in this paper are as follows. First, the change to the six vowel system is almost complete. Second, /ʦ/ is mostly realized as alveolar allophone(=[ʦ]), but in front of /i/, palatal allophone(=[ʧ]) and an alveolar allophone coexist. Third, the palatalized forms and non-palatalized forms coexist. Fourth, vowel palatalization does not occur.

KCI등재

6인공지능 시대의 언어 연구: 주요 문제와 언어학의 역할

저자 : 황화상 ( Hwang Hwa-sang )

발행기관 : 서강대학교 언어정보연구소 간행물 : 언어와 정보사회 45권 0호 발행 연도 : 2022 페이지 : pp. 145-168 (24 pages)

다운로드

(기관인증 필요)

초록보기

Linguists have not played a leading role in natural language processing. And linguistics has not provided enough explanations for the linguistic problems that arise in natural language processing. There is a practical difference between the orientation of linguistics and the orientation of natural language processing. Linguists aim for machines that understand language like humans. In contrast, natural language processing researchers aim for machines that use language as humans do. The natural language processing device inherent in humans is the most perfect. So the surest way to build a machine that uses language is to make it understand language the same way humans understand langauge. For linguists to play a leading role in natural language processing, they have to expand the scope of linguistic research. In particular, linguists must explain how humans understand language. Linguists must also examine the actual process by which we analyze sentences.

KCI등재

7딥러닝 언어모형의 평가와 언어학

저자 : 송상헌 ( Song Sanghoun )

발행기관 : 서강대학교 언어정보연구소 간행물 : 언어와 정보사회 45권 0호 발행 연도 : 2022 페이지 : pp. 169-191 (23 pages)

다운로드

(기관인증 필요)

초록보기

This article addresses how the deep learning-based language models can be evaluated with respect to linguistic knowledge. Building upon the overlook, this article discusses how linguistics can make a substantial contribution to the development of the artificial intelligence systems. As many transformer-based models have been competitively implemented for the last few years, it is required to evaluate the multiple models in a common and reliable way. For this purpose, a wide range of linguistic evaluation metrics have been designed and constructed. The evaluation datasets involve the concepts used in theoretical linguistics, such as syntax, semantics, and pragmatics. The evaluation process follows the guideline used in psycholinguistic experiments. As such, the linguistic knowledge enhances interpretability of the deep leaning-based natural language processing techniques. It is contended that linguistics will play a pivotal role in evaluating and improving the language models in further research.

KCI등재

8중국의 자연어처리 연구 현황과 발전 추세

저자 : 강병규 ( Kang Byeongkwu ) , 박민준 ( Park Minjun )

발행기관 : 서강대학교 언어정보연구소 간행물 : 언어와 정보사회 45권 0호 발행 연도 : 2022 페이지 : pp. 193-231 (39 pages)

다운로드

(기관인증 필요)

초록보기

In this paper, current status and prospects of natural language processing(NLP) research in China will be discussed. With growing importance of NLP, China recently promotes advanced research on NLP like US and Korea. Chinese computational linguistics, commenced with computerization of Chinese character in the mid-1990s, has developed broadly and deeply in various fields such as word segmentation, part-of-speech tagging, named entity recognition, syntax/semantic analysis, information extraction, text summarization/classification, natural language generation/chatbots, etc. As a deep learning approach has been broadly applied to NLP beyond computer vision and speech recognition, Chinese computational linguistics is also facing a massive transformation in terms of research methodology. Following its historical growth, this paper outlines different aspects and macroscopic developments of NLP in China. China's domestic and international status in the field of NLP was examined in quantitative and qualitative measures like publications, citation indexes, academic journals and conferences. This in-depth study on Chinese NLP and its developments at this point suggests an inspirational motivation and direction to Korean NLP society.

KCI등재

9주석 말뭉치의 현황과 전망: 감성 분석 말뭉치를 중심으로

저자 : 김한샘 ( Kim Hansaem )

발행기관 : 서강대학교 언어정보연구소 간행물 : 언어와 정보사회 45권 0호 발행 연도 : 2022 페이지 : pp. 233-262 (30 pages)

다운로드

(기관인증 필요)

초록보기

The annotated corpus, which has been steadily built in the field of language processing, is a learning material that must be continuously supplied to improve the language capacity of artificial intelligence. The importance of the annotated corpus has been further emphasized, and the number of larger and more granular corpora is gradually increasing. Sentiment analysis is a language processing task that determines polarity and classifies it based on positive or negative. Sentiment analysis is a subfield of language processing that requires task-oriented annotation, which has a high practical use and a wide range of applications. The purpose of this paper is to reaffirm the concept and significance of the annotation corpus as a semi-natural language for artificial intelligence. Also, we intend to examine the object and type of annotation in the sentiment analysis corpus by limiting the scope only to sentiment analysis task and focusing on specific cases.

1
권호별 보기

내가 찾은 최근 검색어

최근 열람 자료

맞춤 논문

보관함

내 보관함
공유한 보관함

1:1문의

닫기