키워드 네트워크 분석을 통한 청각장애 관련 연구동향 분석 : 2007~2016

박희정 한국언어치료학회 2018 言語治療硏究 Vol.27 No.4

Purpose : The purpose of this study aimed to investigate the research trends of hearing impairment papers by using a keyword network analysis. Methods : For the analysis, we collected 1372 keywords from 659 research papers published in the domestic journals from 2007 to 2016. For the purpose, frequency analysis, keyword network structural property analysis based on co-keyword network modeling, and keyword network centrality analysis were used. The keyword network analysis was used by UCINET 6.0 program and NetDraw program was then utilized to show the image of some networks. Results : First, the most common keywords were: hearing impairment, cochlear implants, deaf, and student. Second, over the 10-year period, the keyword network of hearing impairment research has a small-world and scale-free network in which a small number of keywords have a tendency to monopolize. As a result of keyword network centrality analysis, keywords with a high degree of centrality, betweenness centrality and closeness centrality were: hearing impairment, cochlear implants, deaf, student, communication, child, university student, sign language, language, program, and inclusive education. The keyword ‘hearing impairment’ was related with all keywords directly. ‘cochlear implants’ was the highest tie-strength keywords. The main keywords were identified as hearing impairment, cochlear implants and deaf. The keyword hearing impairment was related with all keywords directly. However, there were difference between cochlear implants and deaf. Conclusions : This study contributed in understanding the trends of research related with the field of hearing impairment and discussed the prediction of the future research directions. 목적: 본 연구의 목적은 키워드 네트워크 분석을 이용하여 최근 10년(2007~2016) 간 청각장애 관련분야의 연구동향을 살펴보는데 있다. 방법 : 학술연구정보서비스 사이트의 데이터베이스를 활용하여청각장애 관련 논문은 총 659편을 선정하여, 논문에서 제시한 주제어 1372개를 추출하였다. KrKwic 프로그램을 이용하여 10회 이상 출현한 주요키워드의 1 모드 네트워크 행렬을 만들었다. 연구경향을파악하기 위해, 키워드 빈도분석 및 키워드 네트워크의 지적 구조 분석과 중심성 분석을 실시하였다. 키워드 네트워크 분석은 UCINET 6.0을 활용하고, Netdraw 프로그램으로 시각화하였다. 주요 연구결과는 다음과 같다. 결과: 첫째, 키워드 빈도 분석 결과, 10회 이상 출현한 키워드는 41개 중청각장애, 인공와우이식, 농, 학생이 높은 빈도를 나타내었다. 둘째, 청각장애와 관련된 연구는 지난10년 동안 소수의 키워드가 다수의 키워드와 연결된 멱함수 법칙을 따르고 있는 네트워크 구조를보여준다. 중심성 분석 결과, 청각장애, 인공와우이식, 농, 학생, 수어, 아동, 의사소통, 대학생, 난청, 통합교육, 보청기, 프로그램의 연결중심성, 사이중심성, 인접중심성이 모두 높게 나타났다. 청각장애는모든 키워드와 직접적인 연결관계가 있었다. 인공와우는 아동, 보청기, 말지각, 언어, 청능훈련, 심도, 농‘과 연결정도가 높았고, 농은 수어, 중복장애, 농 문화, 한국수어, 인공와우이식, 난청과 높은연결정도를 보여주었다. 결론 : 본 연구를 통해 청각장애 분야 연구의 지적 구조와 최근 동향을전반적으로 살펴볼 수 있었다. 이 결과를 중심으로 청각장애 분야 연구에 있어 부족한 영역을 탐색하고, 향후 새로운 연구 주제 및 방향성에 대해 논의하였다.

키워드 네트워크 분석을 통한 항공우주산업 분야 과학기술의 연구 동향 분석

김희수 ( Hee-su Kim ),김재현 ( Jae-hyun Kim ),황인서 ( In-seo Hwang ),이동훈 ( Dong-hoon Lee ),최원석 ( Won-seok Choi ) 한국항공경영학회 2021 한국항공경영학회지 Vol.19 No.4

연구 동향 분석은 데이터를 이용하여 특정 연구 분야의 동향을 분석하는 방법이다. 본 연구에서는 논문 키워드 네트워크를 통하여 연구 분야의 동향을 살펴본다. 항공우주산업은 주목 받고 있지만 항공우주 산업의 근간이 되는 과학기술에 대한 특징을 데이터에 근거해 분석한 연구는 많이 진행되지 않았다. 본 연구에서는 분석을 위해 2009년~2020년 발행된 항공우주산업 관련 3편의 SCI 저널 (Progress in Aerospace Science, Aerospace Science and Technology, IEEE Transaction On Aerospace and Electronic System) 논문을 선정하였으며, PAS에서 168편, AST에서 844편, TAES에서 2627편 총 3639편의 논문에서 5026개의 키워드를 수집하였다. 수집된 키워드는 정제하여 하나의 논문에서 생성된 키워드들을 링크로 서로 연결하며, 이렇게 구성된 키워드 링크 데이터를 UCINET 툴을 이용해 네트워크를 구성하고 네트워크 분석을 통해 연구 동향 분석을 수행한다. 본 연구를 통해 항공우주 기술 분야의 키워드 네트워크는 평균적으로 하나의 키워드가 8.788개의 키워드와 연결되어있고, 밀도는 0.002 값을 가지며, 클러스터링 계수는 0.451로 매우 낮다. 이는 항공우주산업 기술에서는 항공기 기체, 비행체 제어, 공력, 환경규격, 항공 전자 및 소프트웨어, 인공위성, 추진, 항공기 엔진, 항공기 인증, 발사체, 무인기, 탐지 추적, 광학 등의 다양한 분야가 연구 및 통합되기 때문이다. 또한, 키워드 간 평균 거리가 4.05 값을 가지는데 이는 항공우주기술 분야가 좁은 세상 네트워크(small world network)를 이루고 있으며, 멱함수의 법칙의 특징을 가지는 것을 말한다. 네트워크에서 중심성이 높은 노드는 중요한 위치를 가지며 다른 노드들에 많은 영향력을 끼친다. 연결 중심성(Degree centrality), 매개 중심성(betweenness centrality) 지표를 활용하여 네트워크 내의 중요 키워드들을 분석하고 키워드들의 관계를 살펴봄으로서 항공우주산업의 연구동향을 분석한다. 본 연구를 통해 중요한 키워드로 ‘수신기(Receiver)’, ‘평가(Estimation)’, ‘탐지 및 추적(target track)’, ‘공기역학(aerodynamic)’, ‘궤적(Trajectory)’, ‘전산유체역학(computational fluid dynamic)’, ‘무인항공기(unmanned aerial vehicle)’등의 키워드가 선정되었으며, 이들 키워드는 항공우주산업 과학기술의 근간에서 중요한 키워드로 나타남을 확인할 수 있었다. 본 연구에서는 이 키워드를 중심으로 중요하게 선정된 키워드간의 관계를 살펴본다. Research trend analysis is a method of analyzing trends by using data in a specific research field. In this study, trends in the research field are examined and evaluated with thesis keyword network based on characterize the subject and keyword semantic analyze. Although the aerospace industry has received increasing attention in field of studies, science and technology characteristic analyzing in the aspect of data has not been frequently researched. For this study, 3 SCI journals papers (Progress in Aerospace Science, Aerospace Science and Technology, IEEE Transaction On Aerospace and Electronic System) related to the aerospace industry were selected, which were published from 2009 to 2020. And we collected a total 3639 papers that contains 5026 keywords which are 168 papers form PAS, 844 papers from AST and 2627 papers from TAES. The collected keywords from each paper were connected as one link, and performed network configuration with a linked keyword data set by using UCINET tool. Constructed keyword link data set was analyzed for aerospace trend research. Through this study, we found that a single keyword from aerospace technology field was connected with 8.788 keywords in average, the network keyword density had a value of 0.002 and the clustering coefficient was very low as 0.451. This is because various fields such as aircraft airframe, vehicle control, aerodynamics, environmental standards, avionics and software, satellite, propulsion, aircraft engine, aircraft certification, launch vehicle, unmanned aerial vehicle, detection and tracking and optics are researched and integrated in the aerospace industry technology. In addition, the average distance among keywords has a value of 4.05, which means that the aerospace technology field forms a small world network and has the characteristic of the power law. A node with high centrality in the network link has an important position and has a lot of influence on other nodes. And the research trend of the aerospace industry is analyzed per important keywords in the network using the degree centrality and between centrality indicators and examining the relationship between keywords. From this study, important keywords such as ‘Receiver’, ‘Estimation’, ‘target track’, ‘aerodynamic’, ‘Trajectory’, ‘Computational fluid dynamics’, ‘computational fluid dynamic’ and ‘unmanned aerial vehicle’ were selected. And it confirms that these keywords appeared as important foundation of science and technology in the aerospace industry. For this study also examines the relationship between keywords selected as centering keyword.

키워드 네트워크 분석을 활용한 학생선수 관련 연구동향 분석

강유정,오현정,이병준 한국코칭능력개발원 2023 코칭능력개발지 Vol.25 No.5

이 연구의 목적은 키워드 네트워크 분석을 활용하여 학생선수 관련 국내 논문들의 연구동향을 분석하는 데 있다. 자료수집은 국내 데이터베이스를 이용하여 2000년부터 2023년 3월까지 국내 등재지에 게재된 학생선수 관련 논문 912편을 수집하였다. 자료 분석은 텍스톰과 유시넷을 활용하여 키워드 빈도분석, N-gram 분석, 네트워크 중심성 분석, CONCOR 분석을 실시하였다. 연구 결과는 다음과 같다. 첫째, 전체기간 키워드 네트워크 분석 결과, 가장 많이 출연한 키워드는 학교운동부, 운동선수, 경기력, 진로, 지도자로 확인되었다. 이들 키워드는 중심성 분석 결과에서도 연결 강도가 높을 뿐만 아니라 다른 키워드 사이에 위치하여 매개 역할을 하는 주요 키워드임을 알 수 있었다. N-gram 분석 결과를 통해 학생선수와 학교운동부가 결합하여 가장 많이 등장하는 것으로 확인되었다. CONCOR 분석 결과에서는 비슷한 특성을 가진 키워드를 중심으로 지도자, 학교운동부 정책, 운동수행능력, 진로라는 총 4개의 군집이도출되었다. 둘째, 시기별 키워드 네트워크 분석 결과, 제1시기를 제외하고 제2, 3, 4시기에 가장 많이 출현한 키워드는 학교운동부, 운동선수, 진로, 경기력, 지도자임을 확인하였다. 시기별 연결 중심성의 경우 제1시기부터 제4시기까지 학교운동부와 지도자가 가장영향력이 있는 키워드로 나타났으며, 매개 중심성의 경우 제1시기를 제외하고 나머지 시기에는 학교운동부와 경기력이 다른 키워드사이에 위치하여 중개자 역할을 하는 주요 키워드로 확인되었다. 본 연구의 결과에 근거하여 학생선수와 학교운동부, 학생선수와 지도자, 학생선수의 경기력과 진로 교육의 주제로 연구의 시사점을 논의하였으며, 향후 연구되어야 할 주제와 방향성을 제언했다. The purpose of this study was to analyze the research trends of domestic papers related to student-athletes through keyword network analysis. For data collection, we collected 912 papers related to student athletes published in domestic registration journals from 2000 to March 2023 using domestic databases. Textom and Ucinet were used for conducting data analysis, including keyword frequency analysis, N-gram analysis, network centrality analysis, and CONCOR analysis. The results of the study were as follows. First, as a result of keyword network analysis during the entire period, the keywords that appeared the most were identified as school sports team, athlete, athletic performance, career path, and leader. In the result of centrality analysis, it was found that these keywords not only had high connection strength, but were also major keywords that play a mediating role by being located between other keywords. The results of the N-gram analysis confirmed that the combination of student-athletes and school sports teams appeared most frequently. In a total of four clusters were derived in the result of the CONCOR analysis, which were centered on keywords with similar characteristics: leader, school sports team policy, exercise performance ability, and career path. Second, as a result of keyword network analysis by period, it was confirmed that the keywords that appeared most frequently in the 2nd, 3rd, and 4th periods excluding the 1st period were school sports team, athlete, career path, athlete performance, and leader. In the case of degree centrality by period, from the 1st period to the 4th period, school sports team and leaders appeared as the most influential keywords, In the case of betweenness centrality, except for the first period, in the remaining periods, school sports team and athlete performance were identified as the main keywords acting as mediators as they were located between other keywords. Based on the results of this study, we discussed the implications of research on the subject of student-athletes, school sports teams, student-athletes and leaders, student-athletes performance and career education, and suggested themes and directions to be studied in the future.

이분 네트워크와 다이내믹 네트워크를 활용한 온라인 학습 연구 동향의 종단 분석

사승연,류지훈 부산대학교 교육발전연구소 2024 교육혁신연구 Vol.34 No.1

Purpose: This study sought to investigate longitudinal trends of keywords in online learning related research and changes in the relationship between keywords by year. Method: Through visualization and projection of the bipartite network for 709 papers over five years from 2018 to 2022, the trend of online learning related keywords according to the year was identified. A dynamic network was formed as central keywords to analyze the relationship between the central keywords and the longitudinal changes. Results: From 2020, keywords related to online learning became more diverse. the links with the highest weight in the network from 2020 to 2022 were all links connected to COVID-19. After bipartite network projection, the keywords with a weight of 5 include keywords with high weights in the 2018 network, and there were keywords with a consistently high degree in the dynamic network. In addition, keywords related to online learning include experiment and practice subjects, arts and physical education subjects, and communication-based learning. In the static network, the weight of the link connecting the real-time class and the non-real-time class was the highest. In 2022, comparing the dynamic network with 2020, the weight connecting COVID-19 and the teachers and the degree of the pre-service teacher increased. Lastly, the higher education node has consistently shown a high degree since 2020, and the college student node has noticeably increased their degree with other nodes in 2021 compared to 2020. Conclusion: Based on the results of the study, the authors suggested implications as well as the direction of follow-up study regrading online learning and the selection of author-provided keywords.

여가관련 신문기사의 주제어 연결망 분석

김경식,이경선 한국사회체육학회 2016 한국사회체육학회지 Vol.0 No.64

The purpose of this study was to investigate the change of keywords and to analysis the structure of keyword network on leisure in newspaper articles. Social network naalysis employing NetMiner was performed on leisure articles in national major daily newspapers from 2011 to 2015. The results are as follows. First, p-e riodical leisure keyword has been changed dynamically. The keywords with high level of activity centrality are 40 hours per week, leisure, and workers in 2011, leisure, Chin, aand leisure life in 2012, leisure, leports, an d cycling race in 2013, ocean, rest time, and leisure in 2014, adn finally leisure, MERS, and, ocean in 2015. Ke-y words with high level of medium centrality are leisure, 40 hours per week in 2011, leisure, culture, and Korean in 2012, leisure, sports, and outdoor in 2013, and finally leuisre, free time, Seoul citizen, and MERS in 2015 . Second, leisure keyword network for 5 years appeared in small rgoup networks with grand group network . High appearance frequency keywords are leisure, free time, ocean, and leisure activities. Paired keywords with high concurrent appearance frequency are leisure and ocean, an dleisure and MERS. Keywords for high leve l of centrality activity are leisure, free time, and ocean, and keywords for high medium centrality are leisure, free time, increasing, winter, and ocean. Also minor keywords for liesure showed distribution of inverse function linked with major keywords.

사회 연결망 분석을 통한 기계학회 신뢰성 연구 동향

구자석(Jasuk Koo),박민수(Minsoo Park),이경민(Kyungmin Lee) 대한기계학회 2022 대한기계학회 춘추학술대회 Vol.2022 No.11

Keyword network analysis is the process of investigating network structures in the keyword data using graph theory. In this study, keyword network analysis was conducted on 862 papers presented at the spring conference of the Korean Society of Mechanical Engineers for the past 5 years. Network associations between authors were analyzed using author information. And using the paper title, the frequency of the research keywords and the degree of correlation between the keywords were analyzed. Five large research groups were identified. There are groups where many students gather around a professor, and groups where researchers conduct joint research. Noun and adjective keywords were extracted through morpheme analysis, and the frequency of simultaneous occurrence of a keyword pair in one paper title was confirmed. Keywords were clustered according to network modularity, with the top 5 clusters accounting for 70% of the total data. Through the two-mode keyword network analysis between the researcher and the title keyword, each research field and research networking were confirmed. According to the proposed keyword network analysis procedure, it is possible to identify trends in various research fields from the titles of the paper, identify key researchers, and identify researcher networks.

지적·자폐스펙트럼장애를 위한 음악치료 관련 국내연구의 키워드 의미네트워크 분석

이지경,강이화 대구대학교 한국특수교육문제연구소 2022 특수교육저널 : 이론과 실천 Vol.23 No.4

[Purpose] The purpose of this study was to examine the keywords of domestic research on music therapy for intellectual disabilities (ID) and autism spectrum disorders (ASD), and to clarify the structure by analyzing the semantic network between keywords. [Method] A total of 866 keywords of 238 types were extracted from 197 papers published on the subject of music therapy for ID and ASD in journals between 2000 and 2022. To examine past research trends, the frequency of the appearance of keywords presented in the literature was calculated. By utilizing a UCINET6 program, NetDraw was visualized after analyzing the structural properties, centrality, and ego networks. [Results] In the ID literature, keywords such as "music therapy,'' "child/adolescent with ASD,'' "child/adolescent with ID,'' "social skills,'' "group music therapy,'' "communication skills,'' and "problematic behavior'' were frequently used. In the ASD literature, keywords such as "music therapy,'' "child/adolescent with ASD,'' "child/adolescent with ID,'' "social skills,'' "group music therapy,'' "communication skills,'' and "problematic behavior'' were frequently used. The network had 68 nodes connected by 926 links, and the density was 0.2. In addition, keywords such as "music therapy,'' "child/youth with ASD,'' "child/youth with ID,'' "social skills,'' and "singing activity'' displayed high centrality. In the ego network of "child/adolescent with ID,'' "singing activity,'' "social skills,'' "music intervention,'' "attention concentration,'' and "vocabulary ability'' had high connection strength. In the "child/adolescent with ASD'' ego network, "social skills,'' "stereotyped behaviors,'' "language disorders,'' "communication skills,'' and "problematic behaviors'' had high connection strength. "Music therapy'' showed a strong connection with "child/adolescent with ASD,'' "social skills,'' "music education,'' and "music intervention.'' [Conclusion] This study identified the structure of keywords that were important in music therapy research for ID and ASD in the past, and grasped the trends of past research. Proposals for follow-up studies of comprehensive music therapy studies are also discussed. [목적] 이 연구는 지적·자폐스펙트럼장애를 위한 음악치료 국내연구의 핵심어를 살펴보고, 키워드 간의 의미네트워크를 분석하여 그 구조를 밝히는 것을 목적으로 한다. [방법] 2000년부터 2022년 사이 한국연구재단 등재학술지와 등재후보학술지에 지적·자폐스펙트럼장애를 위한 음악치료를 연구주제로 발행된 논문 197편에서 238종 866개의 키워드를 추출하였다. 과거의 연구 동향을 살펴보기 위해 문헌에 제시된 핵심어의 출현 빈도를 산출하였다. 또한 UCINET6 프로그램을 활용하여 네트워크의 구조적 속성과 중심성, 에고네트워크를 분석한 후 NetDraw로 시각화했다. [결과] 지적·자폐스펙트럼장애를 위한 음악치료 연구에는 “음악치료”, “자폐스펙트럼장애 아동·청소년”, “지적장애 아동·청소년”, “사회성기술”, “그룹음악치료”,“의사소통기술”, “문제행동” 등의 키워드가 빈번하게 사용되었다. 그리고 전체 네트워크에 68개의 노드가 926개의 링크로 연결되고 밀도가 0.2로 산출되었다. 또한 “음악치료”, “자폐스펙트럼장애아동·청소년”, “지적장애 아동·청소년”, “사회성기술”, “노래활동”등의 키워드가 중심성이 높은 핵심어였다. 마지막으로, “지적장애 아동·청소년” 에고네트워크는 “노래활동”, “사회성 기술”, “음악중재”, “주의집중력”, “어휘력”이 연결강도가 높았고, “자폐스펙트럼장애 아동·청소년”에고네트워크는“사회성기술”, “상동행동”, “언어장애”, “의사소통 기술”,“문제행동”이 연결강도가 높았으며, “음악치료”는 “자폐스펙트럼장애 아동·청소년”, “사회성기술”, “음악교육”, “음악중재”와 강한 연결을 보였다. [결론] 이 연구는 과거의 지적·자폐스펙트럼장애를 위한 음악치료 연구에서 중요하게 다루어지는 키워드의 구조를 파악하여 과거 연구의 동향을 파악하고, 음악치료 연구의 후속 연구에 관한 제안점을 논의하였다.

키워드 네트워크 분석을 활용한 영유아 놀이 관련 연구동향 분석

최지은(Jieun Choi) 학습자중심교과교육학회 2019 학습자중심교과교육연구 Vol.19 No.14

본 연구의 목적은 키워드 네트워크 분석을 활용하여 영유아 놀이 관련 연구동향을 분석하는 것이다. 이를 위하여 2015년부터 2019년 5월까지 한국연구재단(KCI) 등재 지와 등재후보지에 게재된 영유아 놀이 관련 학술지 논문의 연도별, 학습지별 동향을 파악하고, 논문에 제시된 키워드에 대한 네트워크 분석을 실시하였다. 분석절차는 분석대상 선정, 키워드 추출과 코딩, 키워드 네트워크 구축, 그리고 키워드 네트워크 분석 및 시각화의 과정으로 진행되었다. 자료처리는 Textom, Ucinet 6.0, NetDraw 프로그램을 활용하였다. 연구결과는 다음과 같다. 첫째, 2015년부터 2019년 사이 총430 편의 영유아 놀이 관련 논문이 게재되었으며, 학술지는 ‘유아교육연구(幼兒敎育硏究)’ 에서 가장 많은 영유아 놀이 관련 논문이 게재된 것으로 나타났다. 둘째, 키워드는 ‘놀이성’이 가장 높은 빈도를 보였으며, ‘또래상호작용’, ‘실외놀이’, ‘교사’, ‘교사유아상 호작용’, ‘또래놀이’ 등의 순으로 높은 빈도를 보였다. 셋째, 연결정도 중심성 분석결과 ‘놀이성’ 지수가 가장 높게 나타났으며, 가장 큰 군집은 ‘영유아 놀이성 지지를 위한 가정기관 소통’이 형성되었다. 본 연구는 전반적인 영유아 놀이 관련 연구동향을 탐색하고, 미래 핵심 연구주제에 대한 방향성을 파악하였다는 데 의의가 있다. The purpose of this study is to examine trends in research on young children’s play using a keyword network analysis. For this purpose, 430 papers were selected from Korean Citation Index journals published from 2015 to 2019 using such keywords as “children’s play” and “young children’s play” in the Research Information Sharing Service. The keyword network analysis was constructed by extracting and coding the keywords that appeared in the selected papers. Textom, Ucinet 6.0, and NetDraw were used to analyze the frequency, degree centrality, and betweenness centrality of the extracted keywords. The results of the keyword network analysis showed that, first, Early Childhood Education Research had the highest frequency of papers examining young children’s play. Second, the keywords/phrases that appeared most frequently in research on young children’s play were “playfulness,” “interaction among peers,” “outdoor play,” “teachers,” and “interaction between teachers and children.” Third, the keywords with high frequencies of occurrence also had high degrees of centrality and betweenness centrality. With regard to the keyword network diagram, it was established that the trends in research on young children’s play are centered on four prominent clusters. Based on these results, it was concluded that future research on young children’s play needs to be examined in further detail and expanded into more specific areas.

언어네트워크 분석을 활용한 「선교와 신학」의 연구동향 분석

김효숙 장로회신학대학교 세계선교연구원 2019 선교와 신학 Vol.47 No.-

In a hyper-connected society, efforts to digitalize various phenomena into networks and scientifically analyze the inherent characteristics in them are becoming increasingly important. The purpose of this study was to explore research trends in Mission and Theology, using a language network analysis that is suitable for the knowledge structure analysis of research papers. The research method was to design a language network analysis by constructing a network based on the co-occurrence of the keywords extracted from the abstracts, exploring the change patterns of the keywords by period, and analyzing the semantic network. To do this, the articles to be analyzed were collected, keywords were extracted and cleaned, and co-occurrence matrices and the network of keywords were developed before NetMiner was used for centrality and community analysis. When analyzing the keywords of Mission and Theology, “mission,” “mission theology,” “Korean church,” “God’s mission,” “ecumenical,” “Gospel,” “the Reformation,” and “culture” were found to be high in frequency and centrality. The research trends of the period shifted from the study of the global keywords, such as “world mission,” to the study of local keywords, such as “Korean church” or “local church.” “God’s mission” or “integral mission” tended to transition to a more contextual and practical discussion to embody the “missional church.” Furthermore, compared with the first half of the study, the second half showed the connections among the keywords in a linear form, and it was recommended that the connection between previous research and the follow-on studies needs to be strengthened. The limitation and delimitation of the study are the facts that the analysis was performed on the keywords, and the problem of bias because of the indexer effect was not excluded. However, this study is meaningful in that semantic networks were analyzed beyond frequency analysis of the research topics and that the change patterns were examined by period and the potential research topics beyond short-term research on the emergence and disappearance of the research topics. 초연결사회에서 다양한 현상들을 네트워크로 정보화하고, 그 속에 내재된 특성들을 과학적으로 분석하려는 노력이 점점 중요해지고 있다. 이 연구는 학술논문의 지식 네트워크 분석에 적합한 언어네트워크 분석을 활용해 「선교와 신학」의 연구 동향을 탐색하고 전망하려는 것이다. 연구방법은 초록에서 추출한 주제어 간의 동시출현을 기준으로 네트워크를 구성하여 주제어의 기간별 변화 양태를 탐색하고, 의미적 연결망을 분석하는 언어네트워크 분석으로 설계하였다. 이를 위해 분석할 논문을 수집하여 주제어를 추출・정제하고, 동시출현 행렬 및 주제어의 네트워크를 개발한 후 NetMiner 프로그램으로 중심성 및 커뮤니티 분석을 실시하였다. 연구결과에 따르면, 「선교와 신학」의 핵심 주제어는 선교, 선교신학, 한국교회, 하나님의 선교, 에큐메니칼, 복음, 종교개혁, 문화 순으로, 빈도와 중심성이 모두 높게 나타났다. 기간별 연구 동향은 글로벌한 ‘세계선교’에 대한 연구가 로컬한 ‘한국교회’ 및 ‘지역교회’를 주제어로 한 연구로 이행하고 있고, ‘하나님의 선교’나 ‘통전적 선교’는 ‘선교적 교회’를 구현하는 보다 상황적이며 실천적인 논의로 이행하는 양상을 띠고 있다. 또한 전반기 연구에 비해 후반기 연구는 주제어 간의 연결이 선형적인 형태를 띠므로 선행연구와 후속연구 간의 연결을 강화할 필요가 있음을 제안하였다. 연구의 한계점이자 제한점은 초록의 주제어를 분석한 것이고, 색인자 효과로 인한 편향의 문제를 배제하지 못했다는 점이다. 그러나 이 연구는 연구주제에 대한 빈도분석을 넘어 의미적 관계를 분석하였으며, 연구주제의 출현과 소멸에 대한 단발적 조사를 넘어 기간별 변화 양태 및 잠재적 연구주제를 논의했다는 점에서 의의가 있다.

사회 연결망 분석을 통한 기계학회 신뢰성 연구 동향

구자석(Jasuk Koo),박민수(Minsoo Park),이경민(Kyungmin Lee) 대한기계학회 2022 대한기계학회 춘추학술대회 Vol.2022 No.11

Keyword network analysis is the process of investigating network structures in the keyword data using graph theory. In this study, keyword network analysis was conducted on 862 papers presented at the spring conference of the Korean Society of Mechanical Engineers for the past 5 years. Network associations between authors were analyzed using author information. And using the paper title, the frequency of the research keywords and the degree of correlation between the keywords were analyzed. Five large research groups were identified. There are groups where many students gather around a professor, and groups where researchers conduct joint research. Noun and adjective keywords were extracted through morpheme analysis, and the frequency of simultaneous occurrence of a keyword pair in one paper title was confirmed. Keywords were clustered according to network modularity, with the top 5 clusters accounting for 70% of the total data. Through the two-mode keyword network analysis between the researcher and the title keyword, each research field and research networking were confirmed. According to the proposed keyword network analysis procedure, it is possible to identify trends in various research fields from the titles of the paper, identify key researchers, and identify researcher networks.