http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
정신건강 진단시스템의 사용 의도에 영향을 미치는 요인에 관한 연구 –소방관 진단회복시스템을 중심으로
설수진,홍수희,김도연,신용태 한국IT정책경영학회 2020 한국IT정책경영학회 논문지 Vol.12 No.6
According to the study of the AI system project, people start to use Chatbot for human mental health. Therefore, we need an empirical study for fire fighters to use an intelligent mental health system without any inconvenience after identifying factors influencing the intention to use the mental health diagnosis system. We found that perceived value, performance, effort expectancy, 5G mobile facilitator factors are important to be considered when implementing the system.
TimeGAN을 활용하여 생성된 배터리 데이터 셋 평가 방안에 대한 연구
설수진(Sujin Seol),김병우(Byeongwoo Kim) 한국자동차공학회 2022 한국자동차공학회 학술대회 및 전시회 Vol.2022 No.11
As environmental issues arise, regulations on vehicles are increasing. Accordingly, the demand for electric vehicles is increasing, and the importance of the battery, which is a key component, is increasing. Recently, SOH estimation research using deep learning is active for battery safety. In order to improve the accuracy of SOH estimation using deep learning, sufficient data is needed both quantitatively and qualitatively. To secure insufficient battery SOH data, TimeGAN was applied to generate new data. RMSE, Euclidean Distance, and DTW were used as indicators to evaluate the generated data. As a result of the analysis, it was confirmed that similar voltage data were generated at similar SOH values.
모바일 웹에서 제조사 및 원격 제어 앱 입력 식별에 관한 실증적 연구
박영식,설수진,이남용 한국IT정책경영학회 2020 한국IT정책경영학회 논문지 Vol.12 No.2
As the number of touch input methods has increased with the increase of mobile device users, research on behavior-based authentication through the radius and strength of touch area with keystroke dynamics in the field of behavior-based authentication has also increased. In general, it has developed into a technology that authenticates users with similar intervals in five identification ranges, and predicts and authenticates even small changes through learning. In this paper, we propose a user device identification method based on the empirical analysis of the values including touch pen and remote input for identification of user touch input for authentication.
정호석,설수진,최주원,김진해 한국IT정책경영학회 2021 한국IT정책경영학회 논문지 Vol.13 No.1
Classified stocks are stocks with special functions, such as the details of property, the exercise of voting rights at the general shareholders' meeting, redemption and conversion, etc., and are determined based on the articles of incorporation. Receive. Even if common stocks are understood as class stocks in order to protect the interests of common shareholders, in this case, it will be difficult to protect the interests of common shareholders. This is because the members of the general shareholders' meeting to approve the amendment of the articles of incorporation and the general shareholders' meeting consisting of common stocks are the same, so even if the latter system is acknowledged, there seems to be no way to neutralize the former resolution. In the end, common shareholders who end up opposing the introduction of new types of stocks that are more advantageous than common stocks should be considered as having no other means of remedy for their rights except by filing an affirmation of the amendment of the articles of incorporation afterwards. In the amendment to the Company Act, in the aspect of defense of management rights, the introduction of the issue limit of unvoting stocks, stocks with veto rights, stocks with appointment and dismissal rights, stocks with multiple voting rights, differential voting stocks, and new stock options were reviewed. It was not introduced into the revised commercial law due to criticism that it could be abused in defense and it did not fit well with our legal system. Therefore, under the revised Commercial Act, a means of defense of management rights using kind stocks should be devised.
개인정보의 비식별화에 따른 기계학습의 예측 정확도 분석 연구
정홍주,이나영,설수진,한경석 한국정보과학회 2020 정보과학회논문지 Vol.47 No.10
The de-identification of personal information is emerging due to the revision of the Personal Information Protection and Personal Information Protection Act. In addition, the use of artificial intelligence and machine learning is becoming a driving force in the Fourth Industrial Revolution. In this paper, we experimentally verify the predictive accuracy of a machine learning decision tree algorithm using de-identified personal information by applying k-anonymity (k=2). The prediction results of the input data are compared to determine the limitations of using de-identified personal information in machine learning. According to the amendment of the Personal Information Protection Act, we propose that when using de-identified personal information in machine learning, the level of personal information de-identification and the analysis algorithm should be considered. 개인정보 보호 및 개인정보 보호법 개정에 따른 개인정보 비식별화 관련 사항이 대두되고 있다. 또한 4차 산업혁명의 원동력으로 인공 지능과 기계학습의 활용이 증대되고 있다. 본 논문에서는 k-익명성(k=2)을 적용한 비식별화 개인정보를 활용하여 기계학습의 의사결정나무 알고리즘으로 예측 정확도를 실험적으로 검증한다. 그리고 입력 데이터의 예측 결과를 비교하여 기계학습에서 비식별화 개인정보를 활용 시 제한 사항을 알아보고자 한다. 개인정보보호법 개정안에 따라 기계학습에 비식별화 개인정보를 사용할 경우, 개인정보 비식별화 수준과 분석 알고리즘을 고려하여 활용해야 한다는 것을 제안한다.
작업 자동화 소프트웨어 API Essential SPECS
홍수희,김재성,설수진,한경석 한국IT정책경영학회 2020 한국IT정책경영학회 논문지 Vol.12 No.3
Integration of services is expanding with the growth of automation software and related industries. Related studies are being actively conducted in academia, and the term is also defined as the Trigger-Action Framework, which is a system that takes over the function of the platform beyond simple software. Supported mobile applications, which were limited to overseas services, are also entering the domestic market and providing infinite convenience to the users of the apps, leading to activation of customers. In this paper, We study the APIs participating in the Trigger-Action Framework and present the appropriate development standard for the service to enhance the usability and suggest the direction of the integrated web service API.
이원석 ( Won Suk Lee ),김동희 ( Dong Hee Kim ),설수진 ( Seol Soo Jin ),신용태 ( Yongtae Shin ) 한국정보처리학회 2020 한국정보처리학회 학술대회논문집 Vol.27 No.2
그동안 4차 산업혁명에 따른 신산업 성장이나 데이터 활용에 걸림돌이 되고 있다는 비판을 받아온 이른바 데이터 3법이 개정됨에 따라 데이터 경제 시대를 본격적으로 맞이하는 마중물이 되리라는 기대가 커지고 있다. 이에 따라 금융권 등 민간영역은 물론이고 중앙부처 등 공공영역에서도 데이터 3법의 개정에 따른 대응 방안을 모색하는 등 분주히 준비하는 상황이다. 따라서 농업분야에서도 데이터 경제 활성화를 목표로 데이터 3법 개정에 따른 대응 방안을 연구하고자 한다.
스마트시티 환경에서 네트워크 스트리밍 연계시스템 특성이 도입 의도에 미치는 요인에 관한 연구 - IT업계 종사자를 중심으로
박상길(Sang-Kil Park),한경석(Kyeong-Seok Han),홍수희(Su-Hee Hong),유현재(Hyun-Jae Yoo),설수진(Su-Jin Seol) 한국디지털콘텐츠학회 2020 한국디지털콘텐츠학회논문지 Vol.21 No.6
In order to improve citizens lives and the competitiveness of cities around the world, smart cities are spreading as a future city model that can realize the balance and harmony of social, economic, and environmental dimensions on the basis of information and communication technology. Therefore, this study intends to understand the factors that influence the intention to introduce the characteristics of the network streaming linkage system necessary for the construction of a smart city. To this end, a network streaming characteristic factor was selected in a smart city environment, and a research model was constructed by applying the TOE framework, technology acceptance model (TAM) theory, and innovation diffusion theory (IDT). To verify the research model and hypothesis, we conducted an empirical survey of IT industry workers. Through this, we verified what factors influenced the intention to introduce Smart Citys network streaming connection.