• 공학교육연구
• /
• 제22권6호
• /
• pp.21-27
• /
• 2019

In the last few years, interest in computer science education has increased significantly. The curriculum is being revised to introduce computer science. Although interest has focused on coding as the main subject, in fact the computer science includes much more than coding. It engages people in being creative with technology as well as understanding the fundamental principles of computer science. Therefore, it is important to consider the curriculum to provide a foundation by teaching and learning computer science. The curriculum is required the development of courses to teach computer science for non-majors in general education. To think like a computer scientist on the knowledge of computer science is computational thinking. In order to maximize the effectiveness of teaching and learning for computational thinking, various teaching methods and supplementary learning materials, and activities should be developed and provided.

• 공학교육연구
• /
• 제18권4호
• /
• pp.66-75
• /
• 2015

Commonly it is perceived that it is difficult to teach engineering design and satisfy accreditation criteria in computer science and engineering related majors. However, since engineering design in these disciplines are a product design, it is more adaptable than other majors which are process design oriented. This paper shows that although there are subtle differences with conventional engineering design, engineering design education in computer science and engineering can be effectively done. This paper concentrates on how engineering design related curriculum can be constructed, what should be considered when designing such curriculum, and how engineering design can be taught in individual design courses(excluding introductory and capstone courses) based on case experience.

• 공학교육연구
• /
• 제22권6호
• /
• pp.3-11
• /
• 2019

To evaluate the suitability of UML (Unified Modelling Language) in terms of educating novice software engineers, we have conducted a survey study with 102 students who were enrolled in a mandatory course for sophomores in computer science and engineering department. The results indicate that UML 1) has high satisfaction rate among software novices as an education material, 2) improves computational thinking and problem-solving skills, and 3) increases the efficacy of learning when the education program is implemented in the order of UML, programming practice, and physical computing. This paper suggests the applicability of UML as a new education paradigm in software education.

• 공학교육연구
• /
• 제19권2호
• /
• pp.60-69
• /
• 2016

This paper deals with the selection method of effective engineering design problem based on LEGO Mindstorm NXT for basic design education. By YouTube case study of various LEGO-based engineering designs for olympic sports, performance criteria have been developed including programming complexity, structural complexity, sensor/actuator complexity and variety of game operation. Programming complexity includes range of programming code length and possible program variety. Structural complexity includes variety of structural elements such as length, shape, weight, and volume to overcome design restrictions. Sensor/actuator complexity includes variety of sensor used and number of possible actuator assemblies. Variety of game operation includes game complexity and required creativity to make LEGO robots. Based on these performance criteria, four representative sports were selected as the candidates for effective engineering design problem. Finally, feasibility and attributes of each candidate were verified by real implementation examples.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2015년도 추계학술발표대회
• /
• pp.1134-1137
• /
• 2015

'영화 시나리오를 기반으로 영화를 어떻게 추천할 수 있는가'에서 본 논문에서는 전통적인 사회관계망 분석 지표 중 그래프의 평균 길이와 평균 군집도 그리고 밀도를 이용하여 3차원의 데이터 집합을 산출했고, 산출한 데이터 집합을 기반으로 k-means 군집화 알고리즘을 활용하여 각 k 값에 따른 영화를 추천해보았다. 그 결과 기타 여느 추천들과 다른 추천결과를 도출해냈다.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2015년도 춘계학술발표대회
• /
• pp.716-719
• /
• 2015

'영화 속에 등장하는 주연들은 어떤 기준으로 선정되는가'에서 본 논문에서는 두 가지 방법을 활용하여 주연들을 추출해보았다. 그 결과 가중치 연결 중심도를 이용한 검출 방법이 공식적인 주연급 등장인물과 일치한다는 것을 도출해냄.

• ETRI Journal
• /
• 제46권4호
• /
• pp.660-670
• /
• 2024

The performance of face recognition (FR) has reached a plateau for public benchmark datasets, such as labeled faces in the wild (LFW), celebrities in frontal-profile in the wild (CFP-FP), and the first manually collected, in-the-wild age database (AgeDB), owing to the rapid advances in convolutional neural networks (CNNs). However, the effects of faces under various fine-grained conditions on FR models have not been investigated, owing to the absence of relevant datasets. This paper analyzes their effects under different conditions and loss functions using K-FACE, a recently introduced FR dataset with fine-grained conditions. We propose a novel loss function called MixFace, which combines classification and metric losses. The superiority of MixFace in terms of effectiveness and robustness was experimentally demonstrated using various benchmark datasets.

• 공학교육연구
• /
• 제17권4호
• /
• pp.95-101
• /
• 2014

As other major, students have difficulties to be successfully adapted to computer science major without some interest and skill to computer programming. In this paper, we try to find successful factors for computer science major freshmen students to computer programming. The factors we focused are programming experience before college entrance, taking liberal arts/natural science courses at high school, application motivations to computer major, existence of mentors, satisfaction to his/her computer department, student's holland job aptitude code. After analysis, taking liberal arts/natural science courses at high school, satisfaction to computer department, some holland job aptitude code are significant to their successful adaptation to computer major. Also, we found the holland job apptitude code is closely related to student's satisfaction to their major for engineering students including computer science students. Our analysis results will be a suggestion for designing computer science education program with students who enters college without some aptitude or preparation to his major.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2016년도 추계학술발표대회
• /
• pp.542-545
• /
• 2016

컴퓨터공학 분야의 논문 정보를 다루고 있는 대표적인 사이트인 DBLP의 연구 동향을 알아보기 위해 본 논문에서는 약 300만개 이상의 논문 서지정보 가져와 분석했다. IT용어 사전을 만들고 각 논문의 제목과 초록에 포함된 주제어를 추출해 분석을 위한 고차원의 행렬을 만들고, k-평균 군집화 기법을 활용하여 1960년도부터 2010년도까지 총 60여 년간의 연대별 주제어 출현 패턴을 분석함으로써 흥미로운 결과를 도출해 냈다.

• Journal of information and communication convergence engineering
• /
• 제21권4호
• /
• pp.287-293
• /
• 2023

With the rapid development of domestic and international over-the-top markets, a large amount of video content is being created. As the volume of video content increases, consumers tend to increasingly check data concerning the videos before watching them. To address this demand, video summaries in the form of plot descriptions, thumbnails, posters, and other formats are provided to consumers. This study proposes an approach that automatically generates posters to effectively convey video content while reducing the cost of video summarization. In the automatic generation of posters, face recognition and clustering are used to gather and classify character data, and keyframes from the video are extracted to learn the overall atmosphere of the video. This study used the facial data of the characters and keyframes as training data and employed technologies such as DreamBooth, a text-to-image generation model, to automatically generate video posters. This process significantly reduces the time and cost of video-poster production.