• Journal of the Society of Disaster Information
• /
• v.18 no.1
• /
• pp.91-103
• /
• 2022

Purpose: After COVID-19, changes in the educational ecosystem take place and fire service academy education system have shifted from face-to-face into non fact-to-face. So, the educational effect of fire officials is decreased and the satisfaction level is also decreased. In this study, we want to examine the current status of non-face-to-face real-time remote education and supplement the problems to improve the educational methods, the educational environment, etc. Method: This study is an independent variable that affects non-face-to-face real-time remote education, consisting of education system environment, self-efficacy of computers, contents (education contents, structure, design, etc.), and proper interaction. A dependent variable was selected with satisfaction for non-face-to-face real-time remote education. In addition, it was selected and analyzed as an active property of learning motivation and learning attitude as control variables. Result: The better the content and the more active the learning motivation and the attitude toward learning, the higher the satisfaction of non-face-to-face real-time remote education, and the more active the learning motivation and the attitude toward learning, the more positive the computer self-efficacy and the satisfaction of learning Conclusion: In order to increase the satisfaction of non-face-to-face real-time education due to COVID-19, education designers or professors need to provide non-face-to-face education contents that can increase the aggressiveness of their learning motivation and learning attitude, and to increase the satisfaction of education for learners by increasing computer self-efficacy through pre-education of non-face-to-face education systems.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.33 no.6
• /
• pp.1021-1031
• /
• 2023

State-sponsored cyber attacks are highly impactful because they are carried out to achieve pre-planned goals. As a defender, it is difficult to respond to them because of the large scale of the attack and the possibility that unknown vulnerabilities may be exploited. In addition, overreacting can reduce the availability of users and cause business disruption. Therefore, there is a need for a response policy that can effectively defend against attacks while ensuring user availability. To solve this problem, this paper proposes a method to collect the number of processes and sessions of defense assets in real time and use them for learning. Using this method to learn reinforcement learning-based policies on a cyber attack simulator, the attack duration based on 100 time-steps was reduced by 27.9 time-steps and 3.1 time-steps for two attacker models, respectively, and the number of "restore" actions that impede user availability during the defense process was also reduced, resulting in an overall better policy.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.58 no.7
• /
• pp.1411-1417
• /
• 2009

In this paper, we consider the design of $H_{\infty}$ control of linear discrete-time systems having no mathematical model. The basic approach is to use Q-learning which is a reinforcement learning method based on actor-critic structure. The model-free control design is to use not the mathematical model of the system but the informations on states and inputs. As a result, the derived iterative algorithm is expressed as linear matrix inequalities(LMI) of measured data from system states and inputs. It is shown that, for a sufficiently rich enough disturbance, this algorithm converges to the standard $H_{\infty}$ control solution obtained using the exact system model. A simple numerical example is given to show the usefulness of our result on practical application.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1995.10b
• /
• pp.71-77
• /
• 1995

In this paper are proposed robust iterative learning control(ILC) algorithms for both linear continuous time-invariant system and linear discrete-time system. In contrast to conventional methods, the proposed learning algorithms are constructed based on both time domain performance and iteration-domain performance. The convergence of the proposed learning algorithms is proved. Also, it is shown that the proposed method has robustness in the presence of external disturbances and the convergence accuracy can be improved. A numerical external disturbances and the convergence accuracy can be improved. A numerical example is provided to show the effectiveness of the proposed algorithm.

• Proceedings of the Korea Institute of Convergence Signal Processing
• /
• 2000.12a
• /
• pp.201-204
• /
• 2000

This paper describes a study on the time-varying control of parameters in learning of the neural network. Elman recurrent neural network (RNN) is used to implement the control of parameters. The parameters of learning and momentum rates In the error backpropagation algorithm ate updated at every iteration using fuzzy rules based on performance index. In addition, the gain and slope of the neuron's activation function are also considered time-varying parameters. These function parameters are updated using the gradient descent algorithm. Simulation results show that the auto-tuned learning algorithm results in faster convergence and lower system error than regular backpropagation in the system identification.

• International Journal of Computer Science & Network Security
• /
• v.21 no.8
• /
• pp.177-181
• /
• 2021

With the recent advances in information and communication technology, online management of students' learning data has become the norm. Research on learning analysis that predicts the near future (in a few years) of students' careers using machine learning methods and state transition models has been widely conducted. It is important for educators to evaluate the comprehension stability of students to prevent a decrease in their comprehension rate and dropouts in the class. In this study, we measured the comprehension process of university students in different types of lectures. Herein, we report on the results of data analysis using time series and data statistics, and consider several educational approaches.

• Journal of Multimedia Information System
• /
• v.8 no.3
• /
• pp.183-190
• /
• 2021

In this paper, we implement a system for a fund recommendation based on the investment propensity and for a future fund price prediction. The investment propensity is classified by scoring user responses to series of questions. The proposed system recommends the funds with a suitable risk rating to the investment propensity of the user. The future fund prices are predicted by Prophet model which is one of the machine learning methods for time series data prediction. Prophet model predicts future fund prices by learning the parameters related to trend changes. The prediction by Prophet model is simple and fast because the temporal dependency for predicting the time-series data can be removed. We implement web pages for the fund recommendation and for the future fund price prediction.

• The Journal of Korea Robotics Society
• /
• v.14 no.1
• /
• pp.1-7
• /
• 2019

Recently, smart factories have attracted much attention as a result of the 4th Industrial Revolution. Existing factory automation technologies are generally designed for simple repetition without using vision sensors. Even small object assemblies are still dependent on manual work. To satisfy the needs for replacing the existing system with new technology such as bin picking and visual servoing, precision and real-time application should be core. Therefore in our work we focused on the core elements by using deep learning algorithm to detect and classify the target object for real-time and analyzing the object features. We chose YOLO CNN which is capable of real-time working and combining the two tasks as mentioned above though there are lots of good deep learning algorithms such as Mask R-CNN and Fast R-CNN. Then through the line and inside features extracted from target object, we can obtain final outline and estimate object posture.

• International journal of advanced smart convergence
• /
• v.7 no.4
• /
• pp.27-39
• /
• 2018

Since the global positioning system (GPS) has been included in mobile devices (e.g., for car navigation, in smartphones, and in smart watches), the impact of personal GPS log data on daily life has been unprecedented. For example, such log data have been used to solve public problems, such as mass transit traffic patterns, finding optimum travelers' routes, and determining prospective business zones. However, a real-time analysis technique for GPS log data has been unattainable due to theoretical limitations. We introduced a machine learning model in order to resolve the limitation. In this paper presents a new, three-stage real-time prediction model for a person's daily route activity. In the first stage, a machine learning-based clustering algorithm is adopted for place detection. The training data set was a personal GPS tracking history. In the second stage, prediction of a new person's transient mode is studied. In the third stage, to represent the person's activity on those daily routes, inference rules are applied.

• Journal of the Korean Society of Industry Convergence
• /
• v.27 no.4_2
• /
• pp.1001-1007
• /
• 2024

Recent deep learning algorithms for object detection in real-time play a crucial role in various applications such as autonomous driving, traffic monitoring, health care, and water quality monitoring. The size of small objects, in particular, significantly impacts the accuracy of detection models. However, data containing small objects can lead to underfitting issues in models. Therefore, this study developed a deep learning model capable of quickly detecting small objects to provide more accurate predictions. The RE-SOD (Residual block based Small Object Detector) developed in this research enhances the detection performance for small objects by using RGB separation preprocessing and residual blocks. The model achieved an accuracy of 1.0 in image classification and an mAP50-95 score of 0.944 in object detection. The performance of this model was validated by comparing it with real-time detection models such as YOLOv5, YOLOv7, and YOLOv8.