• The Journal of the Korea Contents Association
• /
• v.11 no.11
• /
• pp.194-203
• /
• 2011

This study investigated the effectiveness of visual biofeedback simulation training for patients with incomplete spinal cord injuries. Fifteen people with an ASIA impairment scale of C and D, according to the intervention method, were studied. They were randomly divided into two groups, seven people in the experimental group and eight people in the control group. They were studied three times a week for six weeks using a Tetrax Interactive Balance System, for a total of eighteen times, and each test took fifteen minutes. Paired t-tests were used to evaluate the changes before and after intervention. The difference between the groups was compared using an independent t-test. The experimental group showed significantly increased general stability, fall risk index, weight distribution (A, C, D, LEFT, HEEL) and weight distribution index, except for synchronization AC, BD. However, the control group only showed significantly increased weight distribution (A, C, LEFT). In a variation, experimental and control groups showed significantly increased weight distribution (A, B, C, HEEL), general stability, fall risk index and synchronization AD. Consequently, it was found that visual biofeedback simulation training was effective for the balance ability of incomplete spinal cord injury patients, and further studies are required.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.26 no.1
• /
• pp.197-208
• /
• 2016

In this paper, we propose a method based on a convolutional neural network which is one of the deep neural network. So, we convert a malware code to malware image and train the convolutional neural network. In experiment with classify 9-families, the proposed method records a 96.2%, 98.7% of top-1, 2 error rate. And our model can classify 27 families with 82.9%, 89% of top-1,2 error rate.

• Proceedings of the Korea Contents Association Conference
• /
• 2011.05a
• /
• pp.93-94
• /
• 2011

도시지역 대기오염에 의한 건강 위해도 정보는 시민들 삶의 질을 점검하는데 중요하다. 이에 따라 대도시 지역의 여러 오염원들로 부터 배출되는 대기오염물질의 양과 그로 인한 인체 위해도를 시민들이 스마트 폰과 같은 일상적인 통신수단으로 확인할 수 있는 시스템을 설계하고 우리나라 대도시 지역에 적용하는 실험을 하였다. 설계된 시스템에서는 도시 대기로 배출되는 오염물질의 일별 배출량 값들을 산출하고 건강 위해도 인자들을 적용하여 잠재적(potential) 수명 단축 일 수로 정량화한 뒤, 사용자 모델을 통한 적응형 데이터로의 경화량 과정을 거친 후 모바일 기기로 전달되게 된다.

• Annual Conference on Human and Language Technology
• /
• 2019.10a
• /
• pp.161-164
• /
• 2019

본 논문은 BERT 모델을 이용하여 동형이의어의 단어 표현(Word Representation) 차이에 대한 실험을 한다. BERT 모델은 Transformer 모델의 인코더 부분을 사용하여 양방향을 고려한 단어 예측과 문장 수준의 이해를 얻을 수 있는 모델이다. 실험은 동형이의어에 해당되는 단어의 임베딩으로 군집화를 수행하고 이를 Purity와 NMI 점수로 계산하였다. 또한 각 단어 임베딩 사이를 코사인거리(Cosine Distance)로 계산하고 t-SNE를 통해 계층에 따른 변화를 시각화하였다. 군집된 결과는 모델의 중간 계층에서 점수가 가장 높았으며, 코사인거리는 8계층까지는 증가하고 11계층에서 급격히 값이 변하는 것을 확인할 수 있었다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.19 no.5
• /
• pp.1207-1212
• /
• 2015

In this paper, we propose three-dimensional (3D) visualization and recognition techniques of micro-objects under photon-starved conditions using photon counting integral imaging microscopy. To capture high resolution 2D images with different perspectives in the proposed method, we use Synthetic Aperture Integral Imaging (SAII). Poisson distribution which is mathematical model of photon counting imaging system is used to extract photons from the images. To estimate 3D images with 2D photon counting images, the statistical estimation is used. Therefore, 3D images can be obtained and visualized without any damage under photon-starved conditions. In addition, 3D object recognition can be implemented using nonlinear correlation filters. To prove the usefulness of our technique, we implemented the optical experiment.

• Journal of KIISE:Computing Practices and Letters
• /
• v.15 no.3
• /
• pp.226-230
• /
• 2009

Intelligent Tutoring System (ITS) is a procedure of analyzing collected data for teaming, making a strategy and performing adequate service for learners. To perform suitable service for learners, modeling is the first step to collect data from the process of their learning. The model, however, cannot be authentic if collected data can contain learners' inconsistent behaviors or unpredictable learning inclination. This study focused on how to sort normal and abnormal data by analyzing collected data from learners through visualization. A model has been set up to assort unusual data from collected learner's data by using DOLLS-HI which makes possible to diagnose learner's learning propensity based on housing interior learning contents in the experiment. The created model has been confirmed its improved reliability comparing to previous one.

• Journal of Korea Game Society
• /
• v.12 no.6
• /
• pp.95-106
• /
• 2012

Emotion is very important feature in educational situation. Because it has high influence to memory, educational achievement, motivation. This study tried to find out possibility of serious game as emotional engagement tool in educational situation. We did our pilot experiment to elementary school students who are english as second language. In this L2 learning situation, we did our basic experiment with English language learning serious game called 'Word Collectrian". Word Collectrian has some features for emotional engagement. It has interaction for dynamic word visualization, providing context video for word usage, putting visualized word on learner's virtual home. According to experimental result, word Collectrian has possibility for educational achievement and emotional engagement effect.

• The Journal of Korean Association of Computer Education
• /
• v.12 no.5
• /
• pp.1-13
• /
• 2009

Systematic information protection education is necessary to evoke the importance of information protection to the students and to provide them with alternatives to cope with malicious attacks. However, there have been too little development of the methods of systematic education and teaching materials to allow an effective education of information protection in the classrooms today. In this paper, we developed principle-oriented visual learning materials using UCC (User Created Content) for an effective education of information protection and evaluated the effectiveness of our approach with independent sample t-test method after being used in two classes, a test group and a control group. In particular, using UCCs, which have become the most frequently used contents in the Internet, as visual material for the education of information protection, proved to be effective in promoting the interest of students in classes and highly useful in education.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.46 no.2
• /
• pp.106-113
• /
• 2009

Recently, due to many industrial accidents at construction sites, a variety of researches for structural health monitoring (SHM) of buildings are progressing. For real site application of SHM, one of the advanced technologies has blown as wireless sensor networks (WSN). In this paper, we proposed WIVA(WSN Monitoring framework based on 3D Visualization and Augmented Reality in Mobile Devices) system that applies 3D visualization and AR technology to mobile devices with camera based on WSN in order to expand the extent of information can observe. Moreover, we performed experiments to validate effectiveness in 3D and AR mode that utilize WSN data based on IEEE 802.15.4. In real implementation scenario, we demonstrated a fire occurrence test in 3-story building miniature.

• Journal of the Korea Society of Computer and Information
• /
• v.18 no.6
• /
• pp.71-79
• /
• 2013

Metabolic pathway, represented by well-displayed graph, have a complex binding structure, which makes the graphical representation hard to visualize. To apply automatic graph layout techniques to the genome-scale metabolic flow of metabolism domains, it is very important to reduce unnecessary edge crossing on a metabolic pathway layout. we proposed a metabolic pathway layout algorithm based on 3-layer layout. Our algorithm searches any meaningful component existing in a pathway, such as circular components, highly connected nodes, and the components are drawn in middle layer. Then the remaining subgraphs except meaningful components are drawn in upper and lower layer by utilizing a new radial layout algorithm. It reduces ultimately reduced the number of edge crossings. Our algorithm solve the problem that edge crossings exponentially increase as the number of nodes grows.