• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2021.06a
• /
• pp.75-77
• /
• 2021

MPEG 비디오 그룹은 제한된 3D 공간 내에서 움직임 시차(motion parallax)를 제공하면서 원하는 시점(view)을 렌더링(rendering)하기 위한 표준으로 TMIV(Test Model for Immersive Video)라는 테스트 모델과 함께 효율적인 몰입형 비디오의 부호화를 위한 MIV(MPEG Immersive Video) 표준을 개발하고 있다. 몰입감 있는 시각적 경험을 제공하기 위해서는 많은 수의 시점 비디오가 필요하기 때문에 방대한 양의 비디오를 고효율로 압축하는 것이 불가피하다. TMIV 는 여러 개의 입력 시점 비디오를 소수의 아틀라스(atlas) 비디오로 변환하여 부호화되는 화소수를 줄이게 된다. 아틀라스는 선택된 소수의 기본 시점(basic view) 비디오와 기본 시점으로부터 합성할 수 없는 나머지 추가 시점(additional view) 비디오의 영역들을 패치(patch)로 만들어 패킹(packing)한 비디오이다. 본 논문에서는 아틀라스 비디오의 보다 효율적인 부호화를 위해서 패치 내에 생기는 작은 홀(hole)들을 채우는 기법을 제안한다. 제안기법은 기존 TMIV8.0 에 비해 1.2%의 BD-rate 이 향상된 성능을 보인다.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• 2019.11a
• /
• pp.233-234
• /
• 2019

인천시는 최근 해양재난 및 선박사고 해결을 위해 연안 중소형 안전운행 시스템 실증사업을 기획·추진하고 있으며 실증사업을 통해 선박충돌경고 등 4가지 서비스를 제공할 예정이다. 사업 관련 정책 결정 및 사업예산 근거 확보 등 사업의 지속가능성을 위해서는 서비스 효과평가를 통해 사업이 사전에 설정한 목표를 충분히 달성했는지 여부를 평가하고 필요시 추가적인 보완조치 등을 수행하여야 한다. 본 연구에서는 효과평가 정의, 과정, 체계 및 모형 등 효과평가 관련 이론적 접근을 수행하였다. 또한 관련 선행연구와 효과평가 사례분석 등을 통해 효과평가 모델(안)을 제시하였다. 향후 연구 관계자 및 서비스 체험자 등의 의견과 서비스로 인해 생성되는 Data 분석 등을 통해 효과평가 기준을 개발하고 사업에 적용할 예정이며 이를 통해 해양사고 예방 등 사업 목표 달성에 기여할 것으로 기대된다.

• Annual Conference of KIPS
• /
• 2008.05a
• /
• pp.761-763
• /
• 2008

최근 온라인교육의 필요성이 높아지고 요구 수준이 커짐에 따라 교육 서비스를 제공하는 시스템의 지능화된 처리능력이 필요하다. 퍼지신경회로망은 각각의 가중치(weight)를 갖는 채널로 연결한 망형태의 계산모델이다. 퍼지신경회로망을 학습시스템에 적용하여 학습자의 문항테스트 결과에서 학습과정을 재설정 할 수 있는 출력 값을 생성한다. 적응형 학습시스템은 퍼지신경회로망을 적용하여 개별화된 강의 코스로 학습을 진행하고 결과의 feedback을 통해 학습자의 최적 커리큘럼을 찾아내는 방법을 구현하였다.

• The Mathematical Education
• /
• v.63 no.2
• /
• pp.187-207
• /
• 2024

In this study, we explored the feasibility of automatically scoring descriptive assessment items using GPT-4 based ChatGPT by comparing and analyzing the scoring results between teachers and GPT-4 based ChatGPT. For this purpose, three descriptive items from the permutation and combination unit for first-year high school students were selected from the KICE (Korea Institute for Curriculum and Evaluation) website. Items 1 and 2 had only one problem-solving strategy, while Item 3 had more than two strategies. Two teachers, each with over eight years of educational experience, graded answers from 204 students and compared these with the results from GPT-4 based ChatGPT. Various techniques such as Few-Shot-CoT, SC, structured, and Iteratively prompts were utilized to construct prompts for scoring, which were then inputted into GPT-4 based ChatGPT for scoring. The scoring results for Items 1 and 2 showed a strong correlation between the teachers' and GPT-4's scoring. For Item 3, which involved multiple problem-solving strategies, the student answers were first classified according to their strategies using prompts inputted into GPT-4 based ChatGPT. Following this classification, scoring prompts tailored to each type were applied and inputted into GPT-4 based ChatGPT for scoring, and these results also showed a strong correlation with the teachers' scoring. Through this, the potential for GPT-4 models utilizing prompt engineering to assist in teachers' scoring was confirmed, and the limitations of this study and directions for future research were presented.

• Journal of the Korean BIBLIA Society for library and Information Science
• /
• v.30 no.2
• /
• pp.33-50
• /
• 2019

Cataloging principles, cataloging rules, and encoding formats should considered LRM acceptance because LRM replaced FRBR as the conceptual model. This study identifies considerations for BIBFRAME acceptance of expression and representative expression attributes in LRM by using literature reviews and expert interviews. Primarily, work in BIBFRAME without expression as entity could map to work and expression of LRM and sustain expression by linking 2 works (work and expression). Second, BIBFRAME must consider association between representative expression attributes and specific expressions whose values can be transferred to the representative expression attributes. Third, representative expression attributes are different according to work types in LRM, and language, media, intended audience, and scale, that can be used as representative expression attributes in BIBFRAME, should be changed in class. Fourth, relation properties should be articulated for expanding networks between expressions originated from work in BIBFRAME. This study analyzes LRM and BIBFRAME by focusing on expression entity and representative expression attributes. More LRM study is needed on cataloging principles and cataloging rules.

• Korean Journal of Plant Taxonomy
• /
• v.37 no.3
• /
• pp.309-321
• /
• 2007

Korea national arboretum has worked with the plant taxonomic society of Korea to make the first fully electronic floristic checklist in Korea. The result is an ever-expanding online plant name index containing scientifically authorative, up-to-date information on the approximately 7,000 taxa including cultivars. With 37 contributing taxonomists, KPNI is the largest collaborative research projects ever assembled in Korea. A comprehensive database model for the taxonomic data from literature and other sources is presented, which was devised for the Korea National Plant Index database project (KPNI). Gwangreung database model is based on an approach using entity-relationsip diagram. It encompasses taxa of all ranks, nothotaxa and hybrid formulae, cultivars, full synonymy, basionyms, Korean name, and other nomenclatural information. Ths paper presents an analysis of KPNI work processes and an overview how we are approaching the construction of Gwangreung databaese model. It can help the system engineers of other biological information systems to develop their database based on the accurate and integrative taxonomic database.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.21 no.5
• /
• pp.71-77
• /
• 2021

Recently, there is a demand for efficient program development of an IoT application support frameworks considering heterogeneous hardware characteristics. In addition, the scope of hardware support is expanding with the development of neuromorphic architecture that mimics the human brain to learn on their own and enables autonomous computing. However, most existing IoT IDE(Integrated Development Environment), it is difficult to support AI(Artificial Intelligence) or to support services combined with various hardware such as neuromorphic architectures. In this paper, we design an AI component abstract model that supports the second-generation ANN(Artificial Neural Network) and the third-generation SNN(Spiking Neural Network), and implemented an autonomous IoT IDE based on the proposed model. IoT developers can automatically create AI components through the proposed technique without knowledge of AI and SNN. The proposed technique is flexible in code conversion according to runtime, so development productivity is high. Through experimentation of the proposed method, it was confirmed that the conversion delay time due to the VCL(Virtual Component Layer) may occur, but the difference is not significant.

• Korean Journal of Cognitive Science
• /
• v.19 no.3
• /
• pp.297-310
• /
• 2008

Since speech acts implies speakers' intentions, it is essential to determine speakers' speech acts if we want to implement an intelligent dialogue system. We propose a two-step model for effectively determining speakers' speech acts. In the first step, the proposed model returns speech act candidates by using a neural network model based on machine learning and a predictivity model based on statistics, respectively. In the second step, using speech act candidates which are returned by the predictivity model, the proposed model filters out speech act candidates which are returned by the neural network model. Then, the proposed model selects a speech act with maximum output value among the unremoved speech act candidates. In the experiment on a schedule management domain, the proposed two-step modeling method showed better precisions than the previous methods only using a machine learning model or a probability model.

• The Transactions of the Korea Information Processing Society
• /
• v.3 no.5
• /
• pp.1346-1353
• /
• 1996

In this paper, we propose a modeling method for the flip-flops and test generation algorithms to detect the faults in the sequential circuits using VHDL in the high-level design environment. RS, JK, D and T flip-flops are modeled using data flow types. The sequence of micro-operation which is the basic structure of a chip-level leads to a control point where varnishing occurs to one of two micro- operation sequence. In order to model the fault of one micro-operation(FMOP) that perturb another micro-operation effectively, the concept of goal trees and some heuristic rules are used. Given a faulty FMOP or fault of control point (FCON), a test pattern is generated by fault sensitization, path sensitization and determination of the imput combinations that will justify the path sensitization. The fault models are restricted to the data flow model in the ARCHITECTURE statement of VHDL. The proposed algorithm is implemented in the C language and its efficiency is confirmed by some examples.

• Journal of Digital Convergence
• /
• v.11 no.3
• /
• pp.407-413
• /
• 2013

For Ubiquitous service, there are some method researched. To IT convergence study tourism the convergence of IT and tourism in recent years has emerged as a discipline in the future. Tourist information is information about tourism products as tourists tourism decision-making needed to say. Information presented information anytime, anywhere, using a contact-type media, mobile and efficient tourist information content and generate content using Smart App store to the database is needed. This paper, by taking advantage of the Smart App Places to generate content and Smart Things to query, modify, search, tourism information, tourism policy and tourists can be analyzed, and the average inclination and these efficient tourism information content and that can be utilizedmodels are proposed. This u-Smart is a tourist information system. Build the biggest advantages of the meta-meta-model in real time by utilizing Smart App disposition of existing tourism information and tourist and tourism rating database. Helps to generate patterned by digital tourism policy tourism information content.