• Informatization Policy
• /
• v.31 no.2
• /
• pp.3-38
• /
• 2024

Hallucination is a significant barrier to the utilization of large-scale language models or multimodal models. In this study, we collected 654 computer science papers with "hallucination" in the abstract from arXiv from December 2022 to January 2024 following the advent of Chat GPT and conducted frequency analysis, knowledge network analysis, and literature review to explore the latest trends in hallucination research. The results showed that research in the fields of "Computation and Language," "Artificial Intelligence," "Computer Vision and Pattern Recognition," and "Machine Learning" were active. We then analyzed the research trends in the four major fields by focusing on the main authors and dividing them into data, hallucination detection, and hallucination mitigation. The main research trends included hallucination mitigation through supervised fine-tuning (SFT) and reinforcement learning with human feedback (RLHF), inference enhancement via "chain of thought" (CoT), and growing interest in hallucination mitigation within the domain of multimodal AI. This study provides insights into the latest developments in hallucination research through a technology-oriented literature review. This study is expected to help subsequent research in both engineering and humanities and social sciences fields by understanding the latest trends in hallucination research.

• Journal of the Korean Society for information Management
• /
• v.41 no.3
• /
• pp.205-247
• /
• 2024

The purpose of this study is to investigate how information use behavior and satisfaction appear by personality type for liberal arts learning among college students, and to propose a customized information service plan that can help college students study in university libraries. To this end, a survey was conducted on 169 university students enrolled in C University. The analysis consisted of demographic characteristics, MBTI personality type, information use behavior, satisfaction, and university library service perception survey. Frequency analysis, cross-analysis, multinomial logistic regression, one-way ANOVA, and hierarchical regression analysis were performed on the collected data using the SPSS 29 statistical program. As a result of the study, first, significant results were found in 'preferred information sources', 'information source consideration factors', and 'information collection patterns' according to personality type. Second, there were statistically significant differences in satisfaction according to personality type in 'system utilization ability', 'data selection ability', and 'the degree of recognition of the usefulness of learning activities'. Third, in the relationship between preferred information sources and satisfaction based on personality types and information use behaviors, there appears to be an inverse relationship when the content includes various topics with a lack of academic depth or expertise. However, the preference for 'social media' is positively correlated with 'satisfaction with search results,' as it provides diverse perspectives and viewpoints in liberal education

• 유학연구
• /
• v.46
• /
• pp.1-24
• /
• 2019

Gocheong Seogee is a Confucian scholar in the mid-Joseon period during which Confucian philosophy is not completely established as a national discipline. He was an authentic Seon-bi who helped boost Confusion philosophy in southern and northern parts of Chungcheong area including Daejeon, Nonsan, Gongjoo, Buyeo, Hongseong, Jochiwon as well as Boryeong and Dangjin. Following Seogee, Jangsaeng Kim, who had studied under Yulgok, strived to firm up Giho Confusion philosophy there. Influenced by his teacher Geeham Lee's pragmatical school whose main principle is loyal to their country and affectionate to their people, Seogee set his heart on living up to the same path as previous holy men had gone through rather than pursuing an overly speculative school only focused on Confucian scriptures. He seems to have put the virtues of pragmatic Confusion scholars into practice in accordance with the way Confucius and Mencius devoted themselves to their study and learning. He carried Hyangyak into effect, built the first Seowon in Chungnam area called Chunghyeon Seowon emulating Zhu Xi's Baeklokdong Seowon, and paid close attention to learning with Zhu Xi's portrait enshrined. Those things attest that he eagerly strive to put Confucianism into practice. Through these efforts, he gave a significant influence on establishing Confusion philosophy in southern and northern parts of Chungcheong area and trained lots of disciples like Leechang Song, Jaemoon Min - the teacher of Choryeo Yootae Lee - and so on. As a result, he laid a big foundation stone to develop Confusion culture in Naepo region. Therefore, his position should seem to be re-evaluated in Korean Confucian history. In a rigidly hierarchical social structure, he studied and interacted alongside with Geeham Lee, Heon Jo, Ikpeel Song, Sik Jo, and others with no thought of their lower social status and without any academic prejudices. These kinds of his mind-set and attitudes seem to be a precious example of embodying real Seon-bi spirit. So, I suggest Seogee's life has a substantial significance in strengthening personality education of the young today and composing media contents of local community.

• Journal of Broadcast Engineering
• /
• v.2 no.2
• /
• pp.182-193
• /
• 1997

In this paper, we propose technical and structural solutions for various problems in two-way digital communication services in CATV networks by actual field tests. The two-way digital communication services include VOD, video conferencing, distance learning, interactive game and internet service. We investigate the quality of T -1 level 0.544 Mbps) upstream digital signal while varying the number of subscribers in three different cells in accordance with ITU-T G.821. Based on this investigation, several ways to cope with the problems identified are suggested.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.15 no.1
• /
• pp.137-143
• /
• 2014

Smart education in the information window, type talent in modern society is changing as talented smart shoes, smart automation creative talent through education, training can be called education revolution. In other words, the advent of smart devices, such as knowledge and information and to actively growing individual customized training paradigm change in the way education and training to reflect this approach. Therefore, in this paper, a smart learning environment based technologies for implementing the system was designed to be the next generation of cloud computing and N screen-based smart education system was studied. From this, educational functions and features in a smart media environment, based on the analysis of the utilization of a smart education system, which maximizes the system design were studied.

• International journal of advanced smart convergence
• /
• v.2 no.2
• /
• pp.1-6
• /
• 2013

Convergence in technology and media makes it possible to augment nurturing by making blended infrastructures with new digital resources available in learning environments at schools. A game-based provision of educationally-valuable content within a collective, virtual environment combined with an analysis of events in the game after it has been played (blended nurture environment) may help to better communicate human values especially where old-fashioned discussion methods fail or do not fulfill the promise of effective, educational institution-based methods of bringing up. In this paper we discuss the use of behavioural metaphors in game-based blended-nurture scenarios for computer games and pedagogical-psychological tests. Upbringing demands an active intervention in the process of children growing up and the courage to place higher expectations on ourselves and our own lifestyles. Decisions and the behaviour of the younger generation are often irrational and the resulting effects can have destructive consequences. Nowadays, games take on the role of modern storytellers. With the help of complex analogies in games, a variety of situations can be depicted in an eye-catching way and later discussed.

• Journal of Korean Academy of Nursing
• /
• v.45 no.3
• /
• pp.337-346
• /
• 2015

Purpose: This study was conducted to develop a mobile web-based pregnancy health care educational program for mothers who were at an advanced maternal age (AMA) and to verify the effects of the program on pregnancy health care. Methods: This program was developed using a web-based teaching-learning system design model and composed of 10 subject areas. This research was a quasi-experimental study using a non-equivalent control group pretest-posttest time serial design and data were collected from April 2 to May 3, 2014. To verify the effects of the program, it was used for 2 weeks with 30 AMA mothers (experimental group). For the control group, a classroom education booklet for pregnant women used with 31 AMA mothers. Results: The experimental group having participated in program had statistically significantly higher scores for knowledge (t=3.76, p <.001), self-efficacy (t=8.54, p <.001), and practice behavior (t=4.88, p <.001) of pregnancy health care, compared to the control group. Conclusion: The results of the program indicate that a Mobile web-based pregnancy health care educational program is effective in meeting the needs of AMA mothers and can be used as the prenatal educational program for AMA mothers and is appropriate as an educational media for theses mothers.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2019.11a
• /
• pp.71-72
• /
• 2019

최근 공공안전 분야에서는 국민의 위험상황을 분석하여 선제적으로 예측을 하여 국민의 안전을 보장하기 위한 요구사항이 대두대고 있다. 또한 스마트폰 및 스마트워치와 같은 고성능 모바일 단말 기기들의 대중화로 인해 해당 기기들에 부착된 다양한 센서 데이터들을 융복합하여 분석할 경우, 수집한 센서 데이터의 잠재적 가치를 안전보장 측면에서 사용할 수 있는 장점이 있다. 본 논문에서는 대인, 대물, 장소에 해당하는 로그 데이터들을 융복합 분석하여 보호대상자의 안전을 지원하는 다중로그 플랫폼 기반 이동경로 분석 기법을 제안한다. 다중로그 플랫폼에서 수집하는 보호대상자의 이동 경로 궤적을 활용하여 과거에 축적된 이동경로 패턴과 비교를 통해 현재 경로가 평소에 이용하던 경로와의 유사도를 추천하게 된다. 해당 이동 경로 분석 시스템은 위치기반 멀티모달 센서 데이터를 융복합 하여 보호대상자의 안전을 보장하는데 기여 할 것으로 예상된다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2019.11a
• /
• pp.41-43
• /
• 2019

Image-to-image 변환에서 인상적인 성능을 보이는 StarGAN 은 모델의 성능에 중요한 영향을 끼치는 adversarial weight, classification weight, reconstruction weight 라는 세가지 하이퍼파라미터의 결정을 전제로 하고 있다. 본 연구에서는 이 중 conditional GAN loss 인 adversarial loss 와 classification loss 를 대치할 수 있는 attribute loss를 제안함으로써, adversarial weight와 classification weight 를 최적화하는 데 걸리는 시간을 attribute weight 의 최적화에 걸리는 시간으로 대체하여 하이퍼파라미터 탐색에 걸리는 시간을 획기적으로 줄일 수 있게 하였다. 제안하는 attribute loss 는 각 특징당 GAN 을 만들 때 각 GAN 의 loss 의 합으로, 이 GAN 들은 hidden layer 를 공유하기 때문에 연산량의 증가를 거의 가져오지 않는다. 또한 reconstruction loss 를 단순화시켜 연산량을 줄인 simplified content loss 를 제안한다. StarGAN 의 reconstruction loss 는 generator 를 2 번 통과하지만 simplified content loss 는 1 번만 통과하기 때문에 연산량이 줄어든다. 또한 이미지 Framing 을 통해 배경의 왜곡을 방지하고, 양방향 성장을 통해 학습 속도를 향상시킨 아키텍쳐를 제안한다.

• Journal of Broadcast Engineering
• /
• v.25 no.2
• /
• pp.192-199
• /
• 2020

Recently, the utilization of the object tracking algorithm based on the deep learning model is increasing. A system for tracking multiple objects in an image is typically composed of a chain form of an object detection algorithm and an object tracking algorithm. However, chain-type systems composed of several modules require a high performance computing environment and have limitations in their application to actual applications. In this paper, we propose a method that enables real-time operation in low-performance computing environment by adjusting the computational process of object detection module in the object detection-tracking chain type system.