한국어 학습자의 학습 성과에 대한 예측 방법 연구

민경모,박진철 서강대학교 언어정보연구소 2018 언어와 정보 사회 Vol.35 No.-

The purpose of this study is to construct a model that predicts the accomplishment of Korean language learning by using the three learner variables such as gender, age, and Hanja(Chinese characters) culture group, and learner's vocabulary and grammar, reading, and writing scores. In this study, we used the model to find out which learner’s characteristic clues are effective in predicting learner’s accomplishment. As a result, it was confirmed that the learning result of the learner can be predicted about 78% with only six characteristic clues. In addition, it was shown through this study that gender and age factors are not effective factors for predicting the learning result, while Hanja culture group and early learning abilities are important predictors of the learning result. And it was confirmed that the probability of completion of the intermediate course of the non-Hanja culture learners was about 1.6 times higher than Hanja culture learners.

2023 한국언어문화교육학회 제36차 춘계 전국학술대회 <주제 토론> 한국언어문화교육의 현장: 현황과 쟁점

민경모,윤영,김수정,조수진,강현자 한국언어문화교육학회 2023 언어와 문화 Vol.19 No.3

.

Deixis의 개념 정립에 대한 일고찰

민경모 한국어의미학회 2012 한국어 의미학 Vol.37 No.-

This study aims to consider the issues which can be raised in subcategorizing of deixis and the terminology concerning deixis. For this, It discusses the way how the notion of deixis has been accepting in Korean linguistic studies. It also discusses the relation between a point of view on deixis and subcategorizing of deixis as well as the terminology concerning deixis in Korean linguistic studies based on the concept of Origo and Dimension. The results are as follows: First, the way to conceptualize deixis in Korean linguistic studies can be classified roughly into two types; one is based on the relation between deixis expression and context of utterance. the other is based on the relation between origo and the referent of deixis expression. Second, subcategorizing of deixis could be affected by how to conceptualize deixis. The status of Social deixis, Person deixis, and Object deixis depends on a point of view about deixis as a result. Third, Korean equivalent to deixis can be used depending on a point of view about deixis differently.

'미시구조 분류와 접근구조'에 대한 토론문

민경모 한국사전학회 2012 한국사전학회 학술대회 발표논문집 Vol.2012 No.2

인공지능 문장 분류 모델 Sentence-BERT 기반 학교 맞춤형 고등학교 통합과학 질문-답변 챗봇 -개발 및 1년간 사용 분석-

민경모,유준희 한국과학교육학회 2024 한국과학교육학회지 Vol.44 No.3

본 연구에서는 오픈소스 소프트웨어와 인공지능 문서 분류 모델인한국어 Sentence-BERT로 고등학교 1학년 통합과학 질문-답변 챗봇을 제작하고 2023학년도 1년 동안 독립형 서버에서 운영했다. 챗봇은Sentence-BERT 모델로 학생의 질문과 가장 유사한 질문-답변 쌍 6개를 찾아 캐러셀 형태로 출력한다. 질문-답변 데이터셋은 인터넷에 공개된 자료를 수집하여 초기 버전을 구축하였고, 챗봇을 1년 동안 운영하면서 학생의 의견과 사용성을 고려하여 자료를 정제하고 새로운질문-답변 쌍을 추가했다. 2023학년도 말에는 총 30,819개의 데이터셋을 챗봇에 통합하였다. 학생은 챗봇을 1년 동안 총 3,457건 이용했다. 챗봇 사용 기록을 빈도분석 및 시계열 분석한 결과 학생은 수업중 교사가 챗봇 사용을 유도할 때 챗봇을 이용했고 평소에는 방과후에 자습하면서 챗봇을 활용했다. 학생은 챗봇에 한 번 접속하여평균적으로 2.1∼2.2회 정도 질문했고, 주로 사용한 기기는 휴대폰이었다. 학생이 챗봇에 입력한 용어를 추출하고자 한국어 형태소 분석기로 명사와 용언을 추출하여 텍스트 마이닝을 진행한 결과 학생은과학 질문 외에도 시험 범위 등의 학교생활과 관련된 용어를 자주입력했다. 학생이 챗봇에 자주 물어본 주제를 추출하고자 Sentence- BERT 기반의 BERTopic으로 학생의 질문을 두 차례 범주화하여 토픽 모델링을 진행했다. 전체 질문 중 88%가 35가지 주제로 수렴되었고, 학생이 챗봇에 주로 물어보는 주제를 추출할 수 있었다. 학년말에학생을 대상으로 한 설문에서 챗봇이 캐러셀 형태로 결과를 출력하는형태가 학습에 효과적이었고, 통합과학 학습과 학습 목적 이외의 궁금증이나 학교생활과 관련된 물음에 답해주는 역할을 수행했음을 확인할 수 있었다. 본 연구는 공교육 현장에서 학생이 실제로 활용하기에 적합한 챗봇을 개발하여 학생이 장기간에 걸쳐 챗봇을 사용하는과정에서 얻은 데이터를 분석함으로써 학생의 요구를 충족할 수 있는 챗봇의 교육적 활용 가능성을 확인했다는 점에 의의가 있다. This study developed a chatbot for first-year high school students, employing open-source software and the Korean Sentence-BERT model for AI-powered document classification. The chatbot utilizes the Sentence-BERT model to find the six most similar Q&A pairs to a student’s query and presents them in a carousel format. The initial dataset, built from online resources, was refined and expanded based on student feedback and usability throughout over the operational period. By the end of the 2023 academic year, the chatbot integrated a total of 30,819 datasets and recorded 3,457 student interactions. Analysis revealed students’ inclination to use the chatbot when prompted by teachers during classes and primarily during self-study sessions after school, with an average of 2.1 to 2.2 inquiries per session, mostly via mobile phones. Text mining identified student input terms encompassing not only science-related queries but also aspects of school life such as assessment scope. Topic modeling using BERTopic, based on Sentence-BERT, categorized 88% of student questions into 35 topics, shedding light on common student interests. A year-end survey confirmed the efficacy of the carousel format and the chatbot’s role in addressing curiosities beyond integrated science learning objectives. This study underscores the importance of developing chatbots tailored for student use in public education and highlights their educational potential through long-term usage analysis.

한·베 병렬말뭉치를 활용한 한국어와 베트남어 문장의 계량적 특성 비교 연구 : 성경 자료를 바탕으로

민경모 한국국제문화교류학회 2023 문화교류와 다문화교육 Vol.12 No.5

This study is part of a parallel corpus-based contrastive study that examines the differences between Korean and Vietnamese focusing on the quantitative characteristics of sentences based on a Korean-Vietnamese parallel corpus using biblical data. In this study, a parallel corpus arranged by clause was constructed, and after processing it into a morphological annotation corpus through morpheme analysis, the frequency of nouns and verbs and the number of corresponding sentences were extracted. Based on this, a comparison of noun and verb frequencies between Korean and Vietnamese was conducted. The results showed that there were no significant differences in the frequencies of nouns and verbs in Korean and Vietnamese sentences, suggesting that both languages are likely to be verb-oriented languages. Additionally, a comparison of sentence correspondences between Korean and Vietnamese revealed differences in sentence structure, particularly in complex sentences. Vietnamese allows for the connection of clauses without specific markers, often requiring contextual understanding of the relationship between clauses. On the other hand, Korean employs connective endings more actively when linking clauses, and this difference in sentence structure was empirically demonstrated through the comparison of sentence correspondences. Furthermore, it was noted that while Korean connective endings can convey the semantic relationships of “parallel”, “addition”, “time”, “background”, “reason”, “contrast”, “condition”, “transition”, and “concession”, Vietnamese may express these relationships with two independent sentences. Thus we argue that Vietnamese native speakers might face challenges in learning Korean, interpreting or translating due to the complexities of selecting appropriate Korean connective endings.

한국어와 인도네시아어의 문장 단위 대응 양상 연구 : 비대칭 문장 대응 관계를 중심으로

민경모 인하대학교 다문화융합연구소 2024 문화교류와 다문화교육 Vol.13 No.3

학문 목적 한국어 교육용 말뭉치 설계 방안 연구

민경모 한국언어문화교육학회 2010 언어와 문화 Vol.6 No.1

In this study, I will look at data gathering methods in data-based studies for KAP and Corpus Development Activities for EAP in overseas, and then a design and criteria of corpus for KAP teaching is discussed in detail. a corpus criteria to propose in this study is as follows. 1) The corpus is to be multi-purposeful. 2) It is to be designed so that it is used to study a gap between registers in addition to study a characteristic of a specific register. 3) It is to be a principled collection of electronic texts. 4) It is to be dynamic and genre-independent so that additional data gathering is possible. 5) The depth and width of data sampling reflects a diversity of registers and a relative availability and importance of usage situation. The draft plan of a corpus design and corpus criteria like this implies the conversion of subjects from academic language into university Language, as well as the conversion of directions from in-register into inter-register.

해외 초‧중등학교의 한국어 교육과정 설계를 위한 기초적 연구

민경모,김선정 한국언어문화교육학회 2010 언어와 문화 Vol.6 No.3

This study aims at building a general curriculum that can be utilized as a reference to the introduction of Korean language education in elementary and secondary schools overseas. In order to understand the current situation of the schools, a case study in Japan and surveys among other regions have been conducted. The results are as follows : First, the introduction point of Korean language education varies in the different educational systems of each country. Second, it is expected that most Korean language learners remain at the beginning level due to limited practical studying of one to two hours per week. Therefore, with regard to the establishment in standards of the ranks, it is required to consider ranks not by educational stages, but by grades. By examining the rating systems in Korea, European nations, and North America, the conclusion reached is that the application of the Common European Framework of Reference for Languages(CEFR) seems to be most applicable to the current Korean education system in overseas schools. The criteria of CEFR such as 'poly-valence' and 'flexibility', consideration on curriculum planning, applications in diverse regions and comparative advantages in terms of the North American ratings system sustains the validity of the CEFR system in Korean education. However, the ranks in the beginning level are subdivided through distributive approach in anticipation of having a number of beginners. As a result, this study designs an education system of nine stages which subdivides the level of beginner, intermediate and advanced in 4-3-2 stages respectively. On the issue of setting the descriptive categories, by comprehensively examining the domestic and overseas Korean language education curriculum, foreign language education curriculum in elementary and secondary schools in Korea and the horizontal axis of CEFR, three large categories of language, knowledge of language and culture are established. Contained within are a total of 9 descriptive categories with subcategories of 4-3-2 respectively.

병렬 말뭉치와 한국어 교육 연구

민경모 한국언어문화교육학회 2019 언어와 문화 Vol.15 No.1

The purpose of this study is to review the issues related to parallel corpus and previous studies using parallel corpus, and also to examine how to use parallel corpus in the studies of Korean language education. This study starts by summing up the concept, types, and the structure of parallel corpus, and examining problems in construction and use. Next, we review the current trends of language research and language education research using parallel corpus, and then examine the trends of studies related to Korean language education, especially focusing on language contrastive studies. Finally, we would like to conclude what to consider when trying to use parallel corpus in studies of Korean language education, and the direction of the Korean language education research using parallel corpus in the future.