• 문화기술의 융합
• /
• 제10권4호
• /
• pp.595-600
• /
• 2024

학습 성과를 예측하거나 학습 경로를 최적화하는 연구 분야에서 기계학습과 같은 인공지능 기술의 사용이 점차 증가하면서 교육 분야의 인공지능 활용은 점차 많은 진전을 보이고 있다. 이러한 연구는 점차 심층학습과 강화학습과 같은 좀 더 고도화된 인공지능 방법으로 진화하고 있다. 본 연구는 학습자의 과거 학습 성과-이력 데이터를 기반으로 미래의 학습 성과를 예측하는 방법을 개선하는 것이다. 따라서 예측 성능을 높이기 위해 Markov Chain 방법을 응용한 조건부 확률을 제안한다. 이 방법은 기계학습에 의한 분류 예측에 추가하여 학습자가 학습 이력 데이터를 분류 예측에 추가함으로써 분류기의 예측 성능을 향상 시키기 위해 사용된다. 제안 방법의 효과를 확인하기 위해서 실증 데이터인 '교구 기반의 유아 교육 학습 성과 데이터'를 활용하여 기존의 분류 알고리즘과 제안 방법에 의한 분류 성능 지표를 비교하는 실험을 수행하였다. 실험 결과, 분류 알고리즘만 단독 사용한 사례보다 제안 방법에 의한 사례에서 더 높은 성능 지표를 산출한다는 것을 확인할 수 있었다.

• 산업경영시스템학회지
• /
• 제39권3호
• /
• pp.100-108
• /
• 2016

The purpose of this study is to present a business strategy plan to increase organizational effectiveness of small and medium-sized enterprises. The research investigated in the level of human resource activity, such as recruitment, education, evaluation, compensation and development for the employees and executives who are working at small and medium-sized enterprises where located in Busan and Gyongnam province. With this, the research carried out actual proof analysis on the level of human resource activity effects on organization effectiveness like job satisfaction and organizational commitment. The following implications can be acquired from the result of multiple regression analysis on the 201 employees of small and medium enterprises. First, small and medium-sized enterprises should carry out human resource management activities and improve research and development capacity to enhance organization effectiveness. Second, in order to improve job satisfaction of the members of small and medium-sized enterprises, the management should concentrate on recruitment activity and reward maintenance management activity and come up with strategies to enhance learning ability and external network ability. Third, in order to enhance organizational commitment of the members of small and medium-sized enterprises, recruitment activity, training activity, and reward maintenance management activity should be carried out and the management should come up with strategies to enhance learning ability and external network ability. In this research, the objective was only to find out antecedents of organization effectiveness, but considering that causality might arise among the antecedents, in the studies hereafter, the verification on the structural relationship of various factors will be needed.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2023년도 학술발표회
• /
• pp.165-165
• /
• 2023

Streamflow prediction is a critical task in water resources management and essential for planning and decision-making purposes. However, the streamflow prediction is challenging due to the complexity and non-linear nature of hydrological processes. The transfer learning is a powerful technique that enables a model to transfer knowledge from a source domain to a target domain, improving model performance with limited data in the target domain. In this study, we apply the transfer learning using the Informer model, which is a state-of-the-art deep learning model for streamflow prediction. The model was trained on a large-scale hydrological dataset in the source basin and then fine-tuned using a smaller dataset available in the target basin to predict the streamflow in the target basin. The results demonstrate that transfer learning using the Informer model significantly outperforms the traditional machine learning models and even other deep learning models for streamflow prediction, especially when the target domain has limited data. Moreover, the results indicate the effectiveness of streamflow prediction when knowledge transfer is used to improve the generalizability of hydrologic models in data-sparse regions.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권1호
• /
• pp.334-349
• /
• 2022

The effectiveness of Wi-Fi networks is greatly influenced by the optimization of contention window (CW) parameters. Unfortunately, the conventional approach employed by IEEE 802.11 wireless networks is not scalable enough to sustain consistent performance for the increasing number of stations. Yet, it is still the default when accessing channels for single-users of 802.11 transmissions. Recently, there has been a spike in attempts to enhance network performance using a machine learning (ML) technique known as reinforcement learning (RL). Its advantage is interacting with the surrounding environment and making decisions based on its own experience. Deep RL (DRL) uses deep neural networks (DNN) to deal with more complex environments (such as continuous state spaces or actions spaces) and to get optimum rewards. As a result, we present a new approach of CW control mechanism, which is termed as contention window threshold (CW_Threshold). It uses the DRL principle to define the threshold value and learn optimal settings under various network scenarios. We demonstrate our proposed method, known as a smart exponential-threshold-linear backoff algorithm with a deep Q-learning network (SETL-DQN). The simulation results show that our proposed SETL-DQN algorithm can effectively improve the throughput and reduce the collision rates.

• International Journal of Computer Science & Network Security
• /
• 제21권5호
• /
• pp.243-255
• /
• 2021

This study aimed to identify the effectiveness of the use of distance-evaluation tools and methods among students with learning difficulties from the teachers' point of view, to achieve this goal. A scale was built, and the psychometric characteristics were validated. It consisted, in its final form, of 17 items distributed on four axes, in addition to three open questions. It was applied to a random sample of (149) teachers of students with learning difficulties in Makkah Region. The results showed that teachers' keenness to encourage students with learning difficulties, so that they would not feel frustrated with the distance learning process. It was also evident that teachers did not use achievement portfolios in the evaluation process. In connection with the appropriate evaluation methods, the majority indicated the use of work sheets and visual evaluation methods that rely on audio and visual skills, such as presenting videos, pictures, audio and games, and applying short objective tests. Among the proposals to improve evaluation methods and tools: Individual evaluation, attention to individual treatment, obligating personal attendance of students to school, splitting the required tasks, and not increasing the skills required to be mastered. As for the obstacles that teachers face: Lack of time, difficulty in communicating with students with distance learning difficulties and problems related to the Internet such as interruption, weakness, or lack of availability.

• 디지털산업정보학회논문지
• /
• 제14권3호
• /
• pp.103-113
• /
• 2018

The purpose of this study supports the establishment of national e-learning policy by analyzing e-learning status and current status of higher education. Enhance the competitiveness of higher education through sharing information between universities. And to improve e-learning quality management. We surveyed the current status of e-learning in 341 universities and questionnaires about e-learning content, e-learning application form, e-learning platform status was surveyed through each school's learning management system. As a result, the infrastructure of e-learning, the rate of platforms secured, and the contents are increasing gradually each year; however, still, not all students can receive the services equally. Dedicated servers and learning management systems were secured by more than 70% of general universities. In the current development status of e-learning content, multimedia, animation, and text forms are gradually decreasing, but video contents are increasing every year. Most of the online contents were used in the e-learning contents by application type, and blended learning, flipped learning, and mooc is not yet actively used since they are still in the beginning stage. Learning analysis techniques should be supported in order to easily use online learning contents such as flipped learning and mooc. We suggest that the effectiveness of e-learning should be measured and the current state of learning analysis for customized learning should be done. This study aims to contribute to the improvement of competitiveness of higher education by sharing information about e-learning among universities as a basis for improvement of e-learning policy. Future tasks are to improve the customized learning environment by adding whether the system environment for learning analysis is provided at the time of the survey.

• International Journal of Control, Automation, and Systems
• /
• 제6권2호
• /
• pp.282-287
• /
• 2008

In distributed mobile robot systems, autonomous robots accomplish complicated tasks through intelligent cooperation with each other. This paper presents behavior learning and online distributed evolution for cooperative behavior of a group of autonomous robots. Learning and evolution capabilities are essential for a group of autonomous robots to adapt to unstructured environments. Behavior learning finds an optimal state-action mapping of a robot for a given operating condition. In behavior learning, a Q-learning algorithm is modified to handle delayed rewards in the distributed robot systems. A group of robots implements cooperative behaviors through communication with other robots. Individual robots improve the state-action mapping through online evolution with the crossover operator based on the Q-values and their update frequencies. A cooperative material search problem demonstrated the effectiveness of the proposed behavior learning and online distributed evolution method for implementing cooperative behavior of a group of autonomous mobile robots.

• 국제학술발표논문집
• /
• The 3th International Conference on Construction Engineering and Project Management
• /
• pp.338-344
• /
• 2009

Organizational learning of construction team has been long addressed in the literatures, but the mechanism of learning and the influence of leader in the team still remain vague. This paper presents a computational model (OLT) depicting the mechanism and the influence of leader in a systemic way. The OLT model is a multi-agent system based on some eloquent propositions proposed in previous researches. The proposed model is preliminarily validated by some toy-problem simulations. In the OLT model, the leader is assigned as a project manager. The results show that a proper leader can effectively improve the learning process and the result-in performance, in which the team learning is mainly affected by both the leader and the majority in a team. Based on our findings, two propositions are concluded accordingly: (1) Learning of a team would be enhanced if a proper leader is assigned; (2) The effectiveness of learning would increase in a team, in which the members retain explorative attitudes.

• 경영정보학연구
• /
• 제10권2호
• /
• pp.49-70
• /
• 2008

교육의 패러다임이 교수자 중심에서 학습자 중심으로 바뀌면서 가상 교육의 환경이 현실에서 벌어지는 교육환경과 많은 차이가 있을 뿐 아니라 인터넷 기반 교육은 학습자들 간의 다양한 상호작용을 지원할 수 있는 장점도 가지고 있다. 본 연구는 이러한 점을 감안하여 E-Learning 학습에 영향을 미치는 결정요인을 추출해서 추출한 요인들이 E-Learning 기대효과와 학습효과에 어떠한 영향을 미치는지를 분석해 보고자 한다. 이와 같은 연구목적을 위하여 E-Learning 학습에 영향을 미치는 요인들은 크게 학습자의 특성과 시스템 지원환경, 교수자의 특성 3개의 요인으로 구성하였다. 학습자의 특성 요인은 사용의 편의성과 자기효능감의 2개의 요인들로 구성하였으며, 시스템의 특성요인은 컨텐츠 구성, 내용의 적정성, 상호작용정도, 매체 풍부성 4개의 요인으로 구성하였다. 마지막으로 교수자의 특성 요인들로는 강의내용, 강의 스타일, 교수자신용도 등 3개의 요인들로 구성하였다. 분석결과 학습자의 특성은 사용자 편의성이 기대효과와 학습효과에 유의한 관계를 가지는 것으로 나타났다. 이는 시간과 공간의 제약을 받지 않고 인터넷을 통한 학습의 편리성이 학습의 성과가 향상되기를 바라는 학습자의 기대가 컸기 때문인 것으로 생각되고, 또한 학습효과도 학습자가 학습에 거는 기대감이 학습효과에 반영된 것이라고 볼 수 있다.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 2003년도 추계 학술대회 학술발표 논문집
• /
• pp.95-98
• /
• 2003

In this paper, an adaptive fuzzy neural control of unknown nonlinear systems based on the rapid learning algorithm is proposed for optimal parameterization. We combine the advantages of fuzzy control and neural network techniques to develop an adaptive fuzzy control system for updating nonlinear parameters of controller. The Fuzzy Neural Network(FNN), which is constructed by an equivalent four-layer connectionist network, is able to learn to control a process by updating the membership functions. The free parameters of the AFN controller are adjusted on-line according to the control law and adaptive law for the purpose of controlling the plant track a given trajectory and it's initial values are off-line preprocessing, In order to improve the convergence of the learning process, we propose a rapid learning algorithm which combines the error back-propagation algorithm with Aitken's $\delta$$\^$2/ algorithm. The heart of this approach ls to reduce the computational burden during the FNN learning process and to improve convergence speed. The simulation results for nonlinear plant demonstrate the control effectiveness of the proposed system for optimal parameterization.