• Proceedings of the Korean Society of Computer Information Conference
• /
• 2020.01a
• /
• pp.117-120
• /
• 2020

본 논문은 인공지능 메이커 교육과 관련한 요소를 논문 네트워크 키워드 분석과 다양한 빅데이터를 종합하여 핵심용어를 선정 후 인공지능 메이커 교육을 시스템 다이내믹스의 Vensim프로그램으로 인과지도(Casual Loop Diagramming)를 구조분석(모델의 구조)하여 예측 결과를 토대로 향후 미래 상황 추출 및 정책 결정 연구에 영향을 기여한다. 연구 결과 인공지능 교육 정책은 추후 인공지능 교육과 메이커 교육을 융합한 교육 관련 산업이 증대할 것으로 예측되며 교육 경쟁력 향상과 창의적 인재 양성, OTT를 이용한 인공지능 교육 콘텐츠 향상으로 학습에 활용성이 증대하게 된다. 또한 인공지능 교육 정책은 프로그래밍 교육으로 연결되어 성장기 학습자들의 사고력과 정서 발달에 도움 되며 다양한 교재 및 기기 등장으로 인한 학습에 다양성 역시 증가할 것으로 예측된다. 학교 차원에서는 교수·연구 지원 활동이 증가하여 수업 전문성을 가진 교사가 늘어나 학교 교육의 질은 확대되고 학부모는 인공지능 교육 정책에 긍정적으로 된다. 시스템 다이내믹스는 구조가 형태를 결정짓는다는 세계관에 기초하여 피드백 루프와 동태적 형태 유형을 파악하며 다양한 가능성이 존재하게 된다. 이는 추후 다양한 연구를 통해 인공지능 교육 정책 인과지도의 확대로 연결될 수 있음을 암시하며 본 논문을 통해 인공지능 교육 연구 확산에 시발점이 되었으면 한다.

• Annual Conference on Human and Language Technology
• /
• 2020.10a
• /
• pp.184-187
• /
• 2020

다중 지문 기계독해는 질문과 여러 개의 지문을 입력받고 입력된 지문들에서 추출된 정답 중에 하나의 정답을 출력하는 문제이다. 다중 지문 기계독해에서는 정답이 있을 단락을 선택하는 순위화 방법에 따라서 성능이 크게 달라질 수 있다. 본 논문에서는 단락 안에 정답이 있을 확률을 예측하는 단락 재순위화 모델과 선택된 단락에서 서술형 정답을 위한 세부적인 정답의 경계를 예측하는 세부 단락 선별 기법을 제안한다. 단락 순위화 모델 학습의 경우 모델 학습을 위해 각 단락의 출력에 softmax와 cross-entroy를 이용한 손실 값과 sigmoid와 평균 제곱 오차의 손실 값을 함께 학습하고 키워드 매칭을 함께 적용했을 때 KorQuAD 2.0의 개발셋에서 상위 1개 단락, 3개 단락, 5개 단락에서 각각 82.3%, 94.5%, 97.0%의 재현율을 보였다. 세부 단락 선별 모델의 경우 입력된 두 단락을 비교하는 duoBERT를 이용했을 때 KorQuAD 2.0의 개발셋에서 F1 83.0%의 성능을 보였다.

• Journal of the Korean Society for information Management
• /
• v.34 no.4
• /
• pp.7-32
• /
• 2017

Recently, deep learning has been rapidly spreading as an innovative machine learning technique in various domains. This study explored the research trends of deep learning via modified ego centered topic citation analysis. To do that, a few seed documents were selected from among the retrieved documents with the keyword 'deep learning' from Web of Science, and the related documents were obtained through citation relations. Those papers citing seed documents were set as ego documents reflecting current research in the field of deep learning. Preliminary studies cited frequently in the ego documents were set as the citation identity documents that represents the specific themes in the field of deep learning. For ego documents which are the result of current research activities, some quantitative analysis methods including co-authorship network analysis were performed to identify major countries and research institutes. For the citation identity documents, co-citation analysis was conducted, and key literatures and key research themes were identified by investigating the citation image keywords, which are major keywords those citing the citation identity document clusters. Finally, we proposed and measured the citation growth index which reflects the growth trend of the citation influence on a specific topic, and showed the changes in the leading research themes in the field of deep learning.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.16 no.1
• /
• pp.43-48
• /
• 2006

The educational system has been changed owing to Web, which is most actively used on internet and has the characteristics of providing suitable environments for implementing constructivism study theory. WBI(Web Based Instruction), web-mediated teaching form for students at a long distance, has the advantages of possible interact between instructors and learners, offering a great variety of learning materials, and overcome the spatiotemporal restriction. This paper focuces on the construction of learning surroundings where the learner-centered, active learning can be done by design and Implementation of web based instruct system providing a sham examination with an item pool system. The web based Self-directed Learning system for word processor qualifying exams on this paper, can be mentioned as a real item pool that the question is not setting each time by the instructors but can be reused by reference on item pool bank, designed the number of question. It helps the learner Self-directed Learning study with evaluation during the web based instruct process and immediate feedback. It also provides the chance to research some similar using keyword. To sum up, this system can amplify the efficiency of study.

• Journal of The Korean Association For Science Education
• /
• v.35 no.4
• /
• pp.565-572
• /
• 2015

The purpose of this paper is to offer an analysis on the research trends of web-based inquiry learning effectiveness in science education, and to present suggestions for future studies. This study compiled data on 43 articles in Korea and international journals. The content analysis of articles published were from academic journals related to science education and educational technology from 2000 to 2014. The results are as follows: Among domestic articles, the participants ranged from school children to high school students. On the contrary, among foreign articles, the participants are centralized on secondary school students; most used experimental studies; most of the studies resulted with web-based inquiry learning in science education showing effectiveness on science learning performance or science inquiry ability; all web-based inquiry learning were designed using different models of teaching and learning, with the result in the case of domestic research, the utilized models refer to the STS learning model, Internet utilization problem-center inquiry learning model, Procedural model, while in the case of overseas research, the utilized models are SCY, IBLE, and TESI model. Implications of the findings are then discussed, which implies considerations for further research related to web-based inquiry learning.

• The Journal of the Convergence on Culture Technology
• /
• v.6 no.2
• /
• pp.449-454
• /
• 2020

This paper proposes a machine learning-based emotion analysis system that detects a user's depression through their SNS posts. We first made a list of keywords related to depression in Korean, then used these to create a training data by crawling Twitter data - 1,297 positive and 1,032 negative tweets in total. Lastly, to identify the best machine learning model for text-based depression detection purposes, we compared RNN, LSTM, and GRU in terms of performance. Our experiment results verified that the GRU model had the accuracy of 92.2%, which is 2~4% higher than other models. We expect that the finding of this paper can be used to prevent depression by analyzing the users' SNS posts.

• KIPS Transactions on Software and Data Engineering
• /
• v.2 no.4
• /
• pp.281-290
• /
• 2013

The IT asset is a core part that supports the management objective of an organization, and the fast settlement of the IT asset fault is very important. In this study, a fault recovery prediction technique is proposed, which uses the existing fault data to address the IT asset fault. The proposed fault recovery prediction technique is as follows. First, the existing fault recovery data were pre-processed and classified by fault recovery type; second, a rule was established for the keyword mapping of the classified fault recovery types and reported data; and third, a machine learning process that allows the prediction of the fault recovery method based on the established rule was presented. To verify the effectiveness of the proposed machine learning process, company A's 33,000 computer fault data for the duration of six months were tested. The hit rate for fault recovery prediction was approximately 72%, and it increased to 81% via continuous machine learning.

• Journal of Digital Convergence
• /
• v.19 no.11
• /
• pp.17-22
• /
• 2021

In this paper, a machine learning-based automatic infographic recommendation system is proposed to improve the existing infographic production method. This system consists of a part that machine learning multiple infographic images and a part that automatically recommends infographics with artificial intelligence only by inputting basic data from the user. The recommended infographics are provided in the form of a library, and additional data can be input by drag & drop method. In addition, the infographic image is designed to be dynamically adjusted according to the size of the input data. As a result of analyzing the machine learning-based automatic infographic recommendation process, the matching success rate for layout and keyword was very high, and the matching success rate for type was rather low. In the future, a study to improve the matching success rate for the image type for each part of the infographic will be needed.

• Korean Journal of Comparative Education
• /
• v.28 no.1
• /
• pp.1-24
• /
• 2018

The objective of the article is to find the research trends and the main traits presented in the keywords on abstracts of research articles of "International Journal of Education Development" from 2002 to 2017. To do this, Text Network Analysis(TNA) was applied targeting 966 papers on the journal and the major research outcomes are as follows. First, the frequency analysis on the keywords showed that the keywords like Administration of education program, Schools and instruction, Regional public administration, Educational support service, Elementary education, and Elementary and secondary school were analyzed more than 100 times and also high in centrality degree. Second, the analysis results of the keywords presented in those research articles by development goal periods showed that several new keywords like Elementary education, Elementary and secondary school, Education quality, Secondary education, Educational planning have emerged frequently after SDGs and these keywords showed high in their centrality analysis. Third, the analysis on education level showed that the keywords like Elementary education, Administration of education program, School children were high in frequency and centrality degree in Elementary level. In secondary level, Schools and instruction, Administration of education program, Academic achievement were high, and in high level, college and university was high, respectively.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.17 no.5
• /
• pp.470-477
• /
• 2016

This study analyzed the research trends concerning the use of smart devices by young children, such as smart phones, tablet PCs, interactive whiteboards and teacher assistant robots, which has begun to be mentioned relatively recently, and attempted to analyze the characteristics of the research trends and provide guidelines for the direction of future research. A search of articles related to the use of electronic media by young children using an Online Search DB revealed a total of 192 research papers, which were analyzed according to the subject of research, teaching-learning method, area of development and area of activity. It was found that the teaching-learning method, teacher education and professionalism were highly prevalent in the subject of research; the education method integrating play activity with literature activity were highly prevalent in the teaching-learning method; language development and social development were highly prevalent in the area of development; and language activity and social activity were highly prevalent in the area of activity.