• Journal of Intelligence and Information Systems
• /
• v.14 no.3
• /
• pp.59-76
• /
• 2008

In this paper, we propose an agent architecture called L-CAA that is quite effective in real-time dynamic environments. L-CAA is an extension of CAA, the behavior-based agent architecture which was also developed by our research group. In order to improve adaptability to the changing environment, it is extended by adding reinforcement learning capability. To obtain stable performance, however, behavior selection and execution in the L-CAA architecture do not entirely rely on learning. In L-CAA, learning is utilized merely as a complimentary means for behavior selection and execution. Behavior selection mechanism in this architecture consists of two phases. In the first phase, the behaviors are extracted from the behavior library by checking the user-defined applicable conditions and utility of each behavior. If multiple behaviors are extracted in the first phase, the single behavior is selected to execute in the help of reinforcement learning in the second phase. That is, the behavior with the highest expected reward is selected by comparing Q values of individual behaviors updated through reinforcement learning. L-CAA can monitor the maintainable conditions of the executing behavior and stop immediately the behavior when some of the conditions fail due to dynamic change of the environment. Additionally, L-CAA can suspend and then resume the current behavior whenever it encounters a higher utility behavior. In order to analyze effectiveness of the L-CAA architecture, we implement an L-CAA-enabled agent autonomously playing in an Unreal Tournament game that is a well-known dynamic virtual environment, and then conduct several experiments using it.

• Journal of Information Technology and Architecture
• /
• v.10 no.3
• /
• pp.371-378
• /
• 2013

Video-demand learning through E-learning continuously increases on these days. However, not all video-demand learning systems can be utilized properly. When students study by education videos not matched to level of their own, it is possible for them to lose interest in learning. It causes to reduce the learning efficiency. In order to solve the problem, we need to develop a recommendation system which recommends customized education videos according the study levels of students. In this paper, we estimate the study level based on the history of students' watching behaviors such as average watching time, skipping and rewinding of videos. In the experimental section, we demonstrate our recommendation system using real students' video watching history to show that our system is feasible in a practical environment.

• Journal of Fisheries and Marine Sciences Education
• /
• v.28 no.2
• /
• pp.582-595
• /
• 2016

Effects of learning variables on academic achievement were major goal of this study. The differences of learning emotions, attitudes, flow and major satisfaction by grades of college students in medical school were proposed to conduct. Participants of this study consisted of 194 students of 1st and 2nd grade plus 121 students of 3rd and 4th grade in medical school. They completed the survey questionnaires composed of learning emotions, attitudes, flow and major satisfaction. Collected data were analyzed by t-test and stepwise multiple regression. Two kinds of results achieved as follows: First, there were differences of negative and positive emotions, and learning attitudes but were found no differences of learning flow and major satisfaction by grades. Second, there were significant effects of learning emotions and attitudes on academic achievement and also found differences of variables that affect academic achievement by grades. Based on these results, we think necessitate of considering learning emotions and behaviors in developing training programs and students support systems for medical school are obliged.

• Educational Technology International
• /
• v.18 no.2
• /
• pp.193-213
• /
• 2017

This study aimed to explore online learning profiles of in-service teachers in South Korea, focusing on video lecture and discussion activities. A total of 269 teachers took an online professional development course for 14 days, using an online learning platform from which web log data were collected. The data showed the frequency of participation and the initial participation time, which was closely related to procrastinating behaviors. A cluster analysis revealed three online learning profiles of in-service teachers: procrastinating (n=42), passive interaction (n=136), and active learning (n=91) clusters. The active learning cluster showed high-level participation in both video lecture and discussion activities from the beginning of the online course, whereas the procrastinating cluster was seldom engaged in learning activities for the first half of the learning period. The passive interaction cluster was actively engaged in watching video lectures from the beginning of the online course but passively participated in discussion activities. As a result, the active learning cluster outperformed the passive interaction cluster in learning achievements. The findings were discussed in regard to how to improve online learning environments through considering online learning profiles of in-service teachers.

• Research in Mathematical Education
• /
• v.3 no.2
• /
• pp.89-98
• /
• 1999

The purpose of this review of literature is to investigate what kinds of research have been done on social construction of mathematics understanding among elementary students in small groups. Only empirical studies were reviewed, and then grouping was done in terms of the purpose of the study. This grouping identified three categories: 1) Social and mathematical norms in mathematics classroom, 2) Teaching productive communication behaviors for active learning in small group, and 3) Participation roles and communication behaviors in different group structure. To enhance social construction of mathematics understanding in small group settings two suggestions are made: the importance of the selection of collaborative tasks or problems and teachers' beliefs about mathematics and the teaching an learning of mathematics.

• Journal of Information Processing Systems
• /
• v.20 no.2
• /
• pp.159-172
• /
• 2024

The number of healthcare products available for pets has increased in recent times, which has prompted active research into wearable devices for pets. However, the data collected through such devices are limited by outliers and missing values owing to the anomalous and irregular characteristics of pets. Hence, we propose pet behavior recognition based on a hybrid one-dimensional convolutional neural network (CNN) and long short- term memory (LSTM) model using pet wearable devices. An Arduino-based pet wearable device was first fabricated to collect data for behavior recognition, where gyroscope and accelerometer values were collected using the device. Then, data augmentation was performed after replacing any missing values and outliers via preprocessing. At this time, the behaviors were classified into five types. To prevent bias from specific actions in the data augmentation, the number of datasets was compared and balanced, and CNN-LSTM-based deep learning was performed. The five subdivided behaviors and overall performance were then evaluated, and the overall accuracy of behavior recognition was found to be about 88.76%.

• Journal of the Korean Data and Information Science Society
• /
• v.14 no.2
• /
• pp.223-235
• /
• 2003

Using internet, most of students are still exposed to unreasonable commercial information and even tend to consumptive behaviors. Various programs have been mobilized to keep it with a blockade system against the noxious information. But guidance is more instructive than blockade in respect of education. Thus, the focus of this study is to induce and motivate their self-directed learning activities with internet guide contents. We develope a learning guidance material as one of the information platforms. Furthermore, we consider the availability of such learning guidance materials through interview and observation. We find that easy and meaningful internet access and utilizing environment influence the children's self-directed learning ability.

• Knowledge Management Research
• /
• v.15 no.4
• /
• pp.57-78
• /
• 2014

Since team performance has become one of the core factors for companies' success, companies are putting every effort to raise team productivity. In this vein, the purpose of this study was to examine the influence of team learning behavior upon team dynamic capabilities, team effectiveness, and to verify the mediating effect of team dynamic capabilities in corporations. 312 employees were randomly selected to participate in an questionnaire survey. The result has shown that the static correlation exists between team learning behavior, team dynamic capabilities, and team effectiveness. Team dynamic capabilities mediated the relationship between team learning behavior and team effectiveness. Based on the findings, the study implies that learning behaviors among team members should be supported in order to improve its outcome, and HR representatives must help to develop dynamic capabilities.

• Journal for History of Mathematics
• /
• v.27 no.5
• /
• pp.365-384
• /
• 2014

The purpose of this study is to assess the changes that occur to pre-service mathematics teachers by using survey on the self-reflection during their education practices. For four weeks of the education practice period, the changes to pre-service teachers are analyzed from teaching and learning perspectives. The teaching perspective is sub-categorized into lesson contents, teaching methods, and evaluation on teaching, and the learning perspective is sub-categorized into monitoring on learning, support for learning and evaluation on learning. The analysis shows that significant changes occur in teaching contents from the teaching perspective and in all the sub-categories from the learning perspective. Based on the analysis, preservice teachers are suggested to utilize self-reflection programs during their education practices to promote their professionalism in teaching.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.52 no.9
• /
• pp.533-539
• /
• 2003

Reinforcement learning (RL) has been widely used as a learning mechanism of an artificial life system. However, RL usually suffers from slow convergence to the optimum state-action sequence or a sequence of stimulus-response (SR) behaviors, and may not correctly work in non-Markov processes. In this paper, first, to cope with slow-convergence problem, if some state-action pairs are considered as disturbance for optimum sequence, then they no to be eliminated in long-term memory (LTM), where such disturbances are found by a shortest path-finding algorithm. This process is shown to let the system get an enhanced learning speed. Second, to partly solve a non-Markov problem, if a stimulus is frequently met in a searching-process, then the stimulus will be classified as a sequential percept for a non-Markov hidden state. And thus, a correct behavior for a non-Markov hidden state can be learned as in a Markov environment. To show the validity of our proposed learning technologies, several simulation result j will be illustrated.