• Annual Conference of KIPS
• /
• 2015.10a
• /
• pp.1736-1739
• /
• 2015

Social learning encourages and enables learners with common interests to communicate and share knowledge with others through social networks. However, social learning suffers a barrier on communication among learners with various la nguage and culture background. Aiming to avoid this barrier, this paper proposes a framework of cross-language s ocial learning system which can involve more learners' participation on the web. With this framework, an illustrati ve example of task-oriented collaborative learning paradigm is elaborated. It is expected that our proposed system can stimulate more learners to share the learning resource for deep discussions as well as to promote the knowled ge innovation.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.36.6-36
• /
• 2001

The problem of robot navigation and control is a complex task. Its complexity and characteristics depends on the characteristics of the environment robot inhabits, robot construction (mechanical abilities to move, sense) and the job the robot is supposed to do. In this paper we propose a hybrid programming approach to mobile robot navigation and control in an indoor environment. In our approach we used declarative, procedural, and object oriented programming paradigms and we utilized some advantages of our distributed computing architecture. The programming languages corresponding to the paradigms we used were C, C++ and Prolog. In the paper we present some details of our mobile robot hardware and software structure, focusing on the software design and implementation.

• Journal of IKEEE
• /
• v.25 no.4
• /
• pp.619-624
• /
• 2021

Vehicle orientation detection is a challenging task because the orientations of vehicles can vary in a wide range in captured images. The existing methods for oriented vehicle detection require too much computation time to be applied to a real-time system. We propose Rotate YOLO, which has a set of anchor boxes with multiple scales, ratios, and angles to predict bounding boxes. For estimating the orientation angle, we applied angle-related IoU with CIoU loss to solve the underivable problem from the calculation of SkewIoU. Evaluation results on three public datasets DLR Munich, VEDAI and UCAS-AOD demonstrate the efficiency of our approach.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.18 no.4
• /
• pp.151-157
• /
• 2023

RPL (IPv6 Routing Protocol for Low-power Lossy Network) is a standardized routing protocol for LLNs (Low power and Lossy Networks) by the IETF (Internet Engineering Task Force). RPL creates routes and builds a DODAG (Destination Oriented Directed Acyclic Graph) through OF (Objective Function) defining routing metrics and optimization objectives. RPL supports a leaf mode which does not allow any child nodes. In this paper, we propose INFRA-RPL which provides a dynamic leaf mode functionality to a leaf node with the mobility. The proposed protocol is implemented in the open-source IoT operating system, Contiki-NG and Cooja simulator, and its performance is evaluated. The evaluation results show that INFRA-RPL outperforms the existing protocols in the terms of PDR, latency, and control message overhead.

• International Journal of Computer Science & Network Security
• /
• v.23 no.9
• /
• pp.186-191
• /
• 2023

Neural Networks are widely used for huge variety of tasks solution. Machine Learning methods are used also for signal and time series analysis, including electrocardiograms. Contemporary wearable devices, both medical and non-medical type like smart watch, allow to gather the data in real time uninterruptedly. This allows us to transfer these data for analysis or make an analysis on the device, and thus provide preliminary diagnosis, or at least fix some serious deviations. Different methods are being used for this kind of analysis, ranging from medical-oriented using distinctive features of the signal to machine learning and deep learning approaches. Here we will demonstrate a neural network-based approach to this task by building an ensemble of 1D CNN classifiers and a final classifier of selection using logistic regression, random forest or support vector machine, and make the conclusions of the comparison with other approaches.

• International conference on construction engineering and project management
• /
• 2009.05a
• /
• pp.464-471
• /
• 2009

Robot and application of IT skill of construction industry are slow comparatively than another thing industry by the feature. This research proposes progress management and real time information gathering through construction automation and RFID focused on steel structure construction. Building for RTPM system, must consider various variables and surrounding situation in construction field and it is the most important and difficult matter that draw right requirement and grasp relation between this requirements to accomplish one suitable task considering these environment. Therefore, in this study analyzes requirement and target for RTPM system based on scenario that is easy to draw requirement and apply this to use case model. Presented method suggests that represent relation between goals and way that refines goal systematically from requirement of RTPM system. And it could express for visualization through the Way that attaches nonfunctional elements of system with system internal goal.

• Annual Conference on Human and Language Technology
• /
• 2023.10a
• /
• pp.644-647
• /
• 2023

대화 시스템에서 대화 상태 추적은 사용자와의 대화를 진행하면서 사용자의 의도를 파악하여 시스템 응답을 결정하는데 있어서 중요한 역할을 수행한다. 특히 목적지향(task-oriented) 대화에서 사용자 목표(goal)를 만족시키기 위해서 대화 상태 추적은 필수적이다. 최근 다양한 자연어처리 다운스트림 태스크들이 사전학습 언어모델을 백본 네트워크로 사용하고 그 위에서 해당 도메인 태스크를 미세조정하는 방식으로 좋은 성능을 내고 있다. 본 논문에서는 한국어 토큰-프리(token-free) 사전학습 언어모델인 KeByT5B 사용하고 종단형(end-to-end) seq2seq 방식으로 미세조정을 수행한 한국어 생성 기반 대화 상태 추적 모델을 소개하고 관련하여 수행한 실험 결과를 설명한다.

• Journal of Digital Convergence
• /
• v.11 no.8
• /
• pp.235-242
• /
• 2013

This study aimed to evaluate the effectiveness of imagery training on upper limb function and activities of daily living in subacute stroke patients. This study included 16 voluntary participants with subacute stroke. Subjects were randomly assigned to either experimental or control group, with 8 in each group. Imagery training group performed imagery training during 30 minutes and then task-oriented training 30 minutes a day, 5 times a week for 4 weeks. Control group performed task-oriented training during 30 minutes during a day, 5 times a week for 4 weeks. Assessments were made using the Wolf Motor Function Test (WMFT) and Fugl-Meyer motor function assessment (FMA) to evaluate the changes of upper function. And modified Barthel Index (MBI) was measured to evaluate the activities of daily living. The results showed that imagery training group was more significant increase than control group in WMFT, FMA, and MBI (p<.05). Small to huge effect sizes of 1.59, 2.02, 0.37 were observed for WMFT, FMA, and MBI, respectively. This study indicated that imagery training may be helpful in improving the upper limb function and activities of daily living for subacute stroke patients, and support the clinical feasibility of the imagery training.

• Science of Emotion and Sensibility
• /
• v.10 no.4
• /
• pp.523-538
• /
• 2007

This study examined effects of gender and empathic emotional expression of an anthropomorphic interface agent on users. assessment of the agent. In addition, it examined effects of gender and emotional expression regardless of whether visual fidelity of the agent. In Study 1, The agents were manipulated by photographs of human face. The agent expressed empathic emotion by making an other-oriented emotional response congruent with another's perceived welfare. Subjects participated in a task with the agent and then they assessed the agent by rating interpersonal assessment scale. The result reported their preference to the female agent. In addition, they tended to make positive assessment to the agent of opposite gender. In the study 2, gender and expressed emotion of the agent with low fidelity was manipulated. Subjects participated in a task with the agent and then they assessed the agent by rating the same interpersonal assessment scale as study 1. The result reported their preference to the female agent. In addition, they preferred the agent expressing empathic emotion to the agent expressing self-oriented emotion or no emotion. Though the agent had low visual fidelity, its gender and expressed empathic emotion could make a significant effect on users' assessment.

• The Journal of the Acoustical Society of Korea
• /
• v.24 no.8
• /
• pp.498-504
• /
• 2005

We introduce a Korean speech recognition platform (ECHOS) developed for education and research Purposes. ECHOS lowers the entry barrier to speech recognition research and can be used as a reference engine by providing elementary speech recognition modules. It has an easy simple object-oriented architecture, implemented in the C++ language with the standard template library. The input of the ECHOS is digital speech data sampled at 8 or 16 kHz. Its output is the 1-best recognition result. N-best recognition results, and a word graph. The recognition engine is composed of MFCC/PLP feature extraction, HMM-based acoustic modeling, n-gram language modeling, finite state network (FSN)- and lexical tree-based search algorithms. It can handle various tasks from isolated word recognition to large vocabulary continuous speech recognition. We compare the performance of ECHOS and hidden Markov model toolkit (HTK) for validation. In an FSN-based task. ECHOS shows similar word accuracy while the recognition time is doubled because of object-oriented implementation. For a 8000-word continuous speech recognition task, using the lexical tree search algorithm different from the algorithm used in HTK, it increases the word error rate by $40\%$ relatively but reduces the recognition time to half.