• Journal of Broadcast Engineering
• /
• v.28 no.3
• /
• pp.302-313
• /
• 2023

This paper aims to develop a framework that can fully automate the quality management of training data used in large-scale Artificial Intelligence (AI) models built by the Ministry of Science and ICT (MSIT) in the 'AI Hub Data Dam' project, which has invested more than 1 trillion won since 2017. Autonomous driving technology using AI has achieved excellent performance through many studies, but it requires a large amount of high-quality data to train the model. Moreover, it is still difficult for humans to directly inspect the processed data and prove it is valid, and a model trained with erroneous data can cause fatal problems in real life. This paper presents a dataset reconstruction framework that removes abnormal data from the constructed dataset and introduces strategies to improve the performance of AI models by reconstructing them into a reliable dataset to increase the efficiency of model training. The framework's validity was verified through an experiment on the autonomous driving dataset published through the AI Hub of the National Information Society Agency (NIA). As a result, it was confirmed that it could be rebuilt as a reliable dataset from which abnormal data has been removed.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2011.11a
• /
• pp.609-612
• /
• 2011

국내 해양 환경에 대한 조사 및 분석 연구는 미흡한 편에 있다. 최근 세계적으로 바다가 자원의 보고로 주목 받으면서 해양 모니터링 기술에 대한 연구가 활발히 진행 되고 있다. 특히 해양 환경을 분석하고 이해하기 위해서는 지속적으로 해양 환경 자료를 수집해야 하나 아직 많은 부분에서 제약 사항으로 남아 있다. 자동화된 해양 환경 자료의 수집과 수집된 자료를 분석하여서 해양재해를 예측하면 기름 유출에 의한 해양오염의 피해, 적조에 의한 수산업의 피해, 해양환경 이변에 의한 수산업 및 재해 피해를 최소화하는데 기여할 수 있다. 본 논문은 해양환경 감시 및 분석 시스템의 모델을 제안한다. 제안 시스템은 해양환경 정보를 자동 수집하여 해양환경을 지능적으로 감시한다. 또한 수집된 해양 자료를 분석하여서 해양 재해를 예측한다.

• Journal of Industrial Convergence
• /
• v.21 no.5
• /
• pp.133-142
• /
• 2023

In recent years, the problem of academic credential falsification is not simply the manipulation of certificates generated by educational institutions, but also the difficulty of trusting automated relationship verification and verification itself, such as falsely recording completion as graduation while still in school. This is due to the lack of sharing of educational background databases among university institutions and the establishment/operation of independent systems that make it difficult to track educational backgrounds. This study designs and implements an E-AV model for academic credential verification centered on university institutions. It summarizes and stores the linked information on the existing academic background in an encrypted database and implements it with web standard technology considering the compatibility and scalability of the existing system. The results of sample data verification show that it improves safety against forgery and complies with storage space and execution performance. This study aims to contribute to the improvement of online verification services such as academic records management in domestic universities.

• Journal of the Korean Geotechnical Society
• /
• v.23 no.11
• /
• pp.27-37
• /
• 2007

Downhole seismic method is very economic and easy to operate because it uses only one borehole and simple surface source to obtain the shear wave velocity profile of a site. In this study, automatic shear wave source was developed for efficient downhole seismic testing. This source is motor-spring type and easy to control. It can lessen the labor of operator and the working time. Moreover, it can provide better and repetitive signals for data interpretation. By combining developed automatic source with automatic receiver system, PC based data acquisition system, advanced managing program, and semi-automatic downhole performing system were constructed. Through comparison test with manual source, advantages of automatic source were verified. Constructed semi-automatic downhole testing system including automatic shear wave source was applied to the soft soil site. The applicability and reliability were verified and the importance of automating testing system for obtaining reliable result was emphasized.

• Journal of the Korea Society of Computer and Information
• /
• v.14 no.4
• /
• pp.119-129
• /
• 2009

In The increasing use of computers and high-speed Internet network has greatly influenced education, causing a veering away from the typical and traditional way of delivering instruction. Specifically, the various kinds of Web-based multimedia technology, the interactive activities on the Internet, and satellite broadcasting technology are accelerating the emergence of a virtual-lectures-based educational model, which transcends time and space. Such virtual lectures make it possible for the entire teaching-learning process to be done in a virtual learning environment, thus giving rise to problem regarding learning guidance, feedback, and appraisal. In this paper, we propose a system for attendance appraisal for learner participation degree based virtual lecture, an appraisal element in virtual learning environments. This appraisal model can set the elements of virtual learning environments in such a way as to reflect in the attendance appraisal of the opened virtual learning environment information regarding the learner's participation in class. In addition, this model motivates the learners to actively participate in the virtual learning environment and to support instructors by accomplishing the activities that are needed for attendance appraisal.

• Journal of the Korea Society of Computer and Information
• /
• v.14 no.2
• /
• pp.1-8
• /
• 2009

Today the use of RFID is extended in medical. circulation market, manufacture, security field, etc. Besides, the book-lending and the attendance management system are on an increasing trend in the use of RFID The developed dormitory management system is composed of a support program, a room management program, a manager program, and a mobile program etc.. In the support program a student(or a manager) can see his(or all member's) entrance and exit history. The room management program decides whether a student can enter the dormitory or not and manages the use of the power after or inquiring a RFID's input tag information into database server. The manager program is able to input and modify a room state and the fixtures information in the room required to operate the room management program, to set up and retrieve a student's entrance and exit information and a room state information. The mobile program enables a manager to check student's identity whether he is in or out, and the manager can know the personal information of being out in detail.

• Korea Trade Review
• /
• v.42 no.3
• /
• pp.1-24
• /
• 2017

This study investigates any possible influences of the fourth industrial revolution on international trade in Korea and suggests strategies to promote export of Korea in order to secure one of the biggest international trade countries. The fourth industrial revolution is the fourth major industrial era since the third industrial revolution in the 18th century which used electronics and information technology to automate production. This can be characterized as a range of emerging technologies that are fusing the physical, digital and biological worlds, and impacting all disciplines, economies and industries. Since this revolution is expected to have effects on international trade as well as whole industrial society, it examines how it may affect international trade of Korea in terms of the subject, the object, markets and forms of international trade. After that, it provides the strategies to promote Korean export in order to overcome the risks around the low economic growth of the recent years and the depressed domestic economy.

• Design Convergence Study
• /
• v.14 no.5
• /
• pp.145-155
• /
• 2015

With the recent development of ICT, diverse ICT-related fields are being developed and undergoing changes. Numerous researches are being actively conducted on the interactions between users, devices, and computers. Such interaction is now being applied to furniture; interactive furniture are on the rise with which humans can interact with their furniture. This study focused on such interactive furniture and purported to give some pointers to future research on the interaction between users and furniture. This study implemented a user diary on furniture use to analyze user tasks. By using affinity diagram analysis ideation workshop, user tasks were identified and analyzed. The analysis revealed that the future direction for interactive furniture should be automation an complexation. This study is significant in the following two aspects: 1) having suggested future directions for interactive furniture, and 2) serving as foundational data for future research on interactive furniture.

• Asia-pacific Journal of Multimedia Services Convergent with Art, Humanities, and Sociology
• /
• v.7 no.11
• /
• pp.235-244
• /
• 2017

This paper aims to design a system for psychological diagnosis in sandbox play by applying rule based expert system. Sandbox play is one of play therapy and it is a technique that can be combined with psychological diagnosis and treatment using sand and various kinds of figures. In this technique, we focus on psychological diagnostic factors and try to implement a system that automatically diagnoses psychological types when input values are given. Therefore, six kinds of language objects are set and the rules are created according to the types of figures, arrangement of figures, and production time in the sand box used as a reference element in the diagnosis method. In this system, it is assumed that the client recognizes the finished sandbox as a sensor device. Then, when the recognized state enters the input value, the rules based on the language object are ignited in order. Through this, the client is diagnosed with one of 26 types of psychology. As a result, the diagnostic process is simplified and automated. Accordingly, a more detailed psychological diagnosis and treatments are provided.

• Journal of the Korean Society for information Management
• /
• v.40 no.2
• /
• pp.137-156
• /
• 2023

As social and political paradigms change, public institution tasks and structures are constantly created, integrated, or abolished. From an effective record management perspective, it is necessary to review whether the previously established record classification schemes reflect these changes and remain relevant to current tasks. However, in most institutions, the restructuring process relies on manual labor and the experiential judgment of practitioners or institutional record managers, making it difficult to reflect changes in a timely manner or comprehensively understand the overall context. To address these issues and improve the efficiency of record management, this study proposes an approach using automation and intelligence technologies to restructure the classification schemes, ensuring records are filed within an appropriate context. Furthermore, the proposed approach was applied to the target institution, its results were used as the basis for interviews with the practitioners to verify the effectiveness and limitations of the approach. It is, aiming to enhance the accuracy and reliability of the restructured record classification schemes and promote the standardization of record management.