• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.4
• /
• pp.501-507
• /
• 2021

Neural networks with differentiable activation functions are differentiable with respect to input variables. We improve the approximation capability of neural networks by using the gradient and Hessian of neural networks to satisfy the differential equations of the problems of interest. We apply differential neural networks to the pricing of financial options, where stochastic differential equations and the Black-Scholes partial differential equation represent the differential relation of price of option and underlying assets, and the first and second derivatives of option price play an important role in financial engineering. The proposed neural network learns - (a) the sample paths of option prices generated by stochastic differential equations and (b) the Black-Scholes equation at each time and asset price. Experimental results show that the proposed method gives accurate option values and the first and second derivatives.

• The Journal of the Convergence on Culture Technology
• /
• v.8 no.4
• /
• pp.37-44
• /
• 2022

The purpose of this study is to investigate how the local community case management council perceives the expected outcome of the council and case management service by participating actors in order to establish the identity and direction of the activities of the council through concept mapping. A total of 12 practitioners from participating organizations freely expressed and shared their opinions about the case management service performance with the local community case management council in a brainstorming manner, producing a total of 42 statements. Through concept mapping, participants were empowered in the decision-making process, and their opinions were not alienated, but were accepted and rationally handled. It can be said that it is important to provide an opportunity for the participants to discuss on an equal footing in the decision-making process. Through this, it can be seen that the agreement between the case management council and the case management service was quickly reached, and the direction for subsequent activities was clearly set.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.8
• /
• pp.1222-1230
• /
• 2022

As the number of signal sources with low detectability by using various modulation techniques increases, research to classify signal modulation methods is steadily progressing. Recently, a Convolutional Neural Network (CNN) deep learning technique using FFT as a preprocessing process has been proposed to improve the performance of received signal classification in signal interference or noise environments. However, due to the characteristics of the FFT in which the window is fixed, it is not possible to accurately classify the change over time of the detection signal. Therefore, in this paper, we propose a CNN model that has high resolution in the time domain and frequency domain and uses wavelet transform as a preprocessing process that can express various types of signals simultaneously in time and frequency domains. It has been demonstrated that the proposed wavelet transform method through simulation shows superior performance regardless of the SNR change in terms of accuracy and learning speed compared to the FFT transform method, and shows a greater difference, especially when the SNR is low.

• Journal of the Korea Convergence Society
• /
• v.12 no.11
• /
• pp.319-328
• /
• 2021

The purpose of this study is to develop core competencies for gifted education by utilizing Delphi survey methods and to present behavioral element selection and scale questions based on the definition of competencies. First, the core competence for gifted education was selected through literature analysis, and the first Delphi survey was conducted to verify that the definition of each competency is suitable for the competency name. Subsequently, through a second Delphi survey, detailed questions were developed and verified by expressing the capabilities required to develop core competencies as behavior elements. Through two rounds of Delphi surveys, eight key competencies were finally selected: attitude and practice willingness, communication and collaboration, information processing and tool utilization, creative problem solving, convergence and application, higher-order inference, community spirit, and learning achievement orientation. This study is meaningful in that it selects core competencies and behavior elements for gifted education that are necessary to pursue goals that meet social needs and it presents tools to measure the degree of competency improvement for gifted education.

• The Journal of the Convergence on Culture Technology
• /
• v.8 no.5
• /
• pp.219-228
• /
• 2022

The purpose of this study is to explore what obstacles adult undergraduate experience in adapting to college life. To this end, in-depth interviews were conducted with 32 adult undergraduate attending colleges in Daegu and Gyeongbuk. For the study, Colaizzi's phenomenological research method was used and analyzed. As a result of the analysis, eight factors such as relation obstacle, bachelor's and curriculum operation obstacle, social recognition obstacle, study ability obstacle, college environment obstacle, economic obstacle, personal disposition obstacle, and temporal obstacle were found. Through the above research results, it was found that the college environment, which is operated mainly by general college students, is expressed as various types of obstacle for adult undergraduate who work and study at various ages and experiences. Based on the derived obstacle factors, it is expected that a follow-up study will be conducted to develop a measurement tool that can empirically explore the obstacle of adult undergraduate to adapt to college life.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.33 no.4
• /
• pp.621-631
• /
• 2023

Adversarial training improves the robustness of deep neural networks for adversarial examples. However, the previous adversarial training method focuses only on the adversarial loss function, ignoring that even a small perturbation of the input layer causes a significant change in the hidden layer features. Consequently, the accuracy of a defended model is reduced for various untrained situations such as clean samples or other attack techniques. Therefore, an architectural perspective is necessary to improve feature representation power to solve this problem. In this paper, we apply an attention module that generates an attention map of an input image to a general model and performs PGD adversarial training upon the augmented model. In our experiments on the CIFAR-10 dataset, the attention augmented model showed higher accuracy than the general model regardless of the network structure. In particular, the robust accuracy of our approach was consistently higher for various attacks such as PGD, FGSM, and BIM and more powerful adversaries. By visualizing the attention map, we further confirmed that the attention module extracts features of the correct class even for adversarial examples.

• The Journal of the Convergence on Culture Technology
• /
• v.9 no.6
• /
• pp.813-825
• /
• 2023

This study was analyzed using a mixed methods approach to clarify the concept of compassionate communication competencies of psychiatric nurses. In the theoretical phase, the literature published from 2000 to 2022 was collected and 38 articles were analyzed. For the fieldwork phase, in-depth interviews were conducted with eight psychiatric nurses from December 1 to December 28, 2022. In the final analysis phase, the dimensions and attributes of psychiatric nurses' compassionate communication competence were identified and conceptualized. Based on the attributes identified in the theoretical and fieldwork phases, the definition of psychiatric nurses' compassionate communication competence was synthesized into five dimensions and 12 attributes. Therefore, psychiatric nurses' compassionate communication competence refers to the skills and abilities of psychiatric nurses to use active listening and empathic skills for effective communication based on compassion and understanding of the target, to be sensitive to the thoughts and feelings of the target, to accurately convey what the target wants to express, to respect the target, and to empower the target.

• Journal of Industrial Convergence
• /
• v.21 no.12
• /
• pp.37-43
• /
• 2023

In today's business landscape, collaboration and interoperability are crucial for organizational success and profitability. However, integrating operations across multiple organizations is challenging due to differing roles and policies in Identity and Access Management (IAM). User-centric identity (UCI) adopts a personalized approach to digital identity management, centering on the end-user for authentication and access control. It provides a decentralized system that ensures secure and customized access for each user. UCI aims to address complex security challenges by aligning access privileges with individual user requirements. This research delves into UCI's ability to streamline resource access amidst conflicting IAM roles and protocols across various organizations. The study presents a UCI-based multi-domain access control (MDAC) framework, which encompasses an ontology, a unified method for articulating access roles and policies across domains, and software services melding with UCI infrastructure. The goal is to enhance organizational resource management and decision-making by offering clear guidelines on access roles and policy management across diverse domains, ultimately boosting companies' return on investment.

• Smart Media Journal
• /
• v.12 no.11
• /
• pp.1-9
• /
• 2023

Stress is a significant issue in modern society, often triggered by external or internal factors that are difficult to manage. When high stress persists over a long term, it can develop into a chronic condition, negatively impacting health and overall well-being. However, it is challenging for individuals experiencing chronic stress to recognize their condition, making early detection and management crucial. Using biosignals measured from wearable devices to detect stress could lead to more effective management. However, there are two main problems with using biosignals: first, manually extracting features from these signals can introduce bias, and second, the performance of classification models can vary greatly depending on the subject of the experiment. This paper proposes a model that reduces bias using convo utional autoencoders, which can represent the key features of data, and enhances generalizability by employing soft voting, a method of ensemble learning, to minimize performance variability. To verify the generalization performance of the model, we evaluate it using LOSO cross-validation method. The model proposed in this paper has demonstrated superior accuracy compared to previous studies using the WESAD dataset.

• Journal of the Korean Society of Earth Science Education
• /
• v.3 no.2
• /
• pp.99-108
• /
• 2010

This study lays its purpose on examining the effects of lessons with the application of drawing tasks on changes in conception among gifted science students. The lesson with the application of drawing tasks means the lesson where students express key concepts regarding lesson subjects in drawings which are then applied to the lessons to develop conception among the learners. This study analyzed the effectiveness of lessons by comparing conception scores before and after experiments between an experiment group with the application of drawing tasks and a control group with normal lessons for the gifted in general for 8 months from March to October, 2008. In addition, the researcher examined how the effectiveness of the developed lessons show differently according to levels of meta-cognition, creative problem-solving abilities, and scientific inquiry skills among the gifted students. The results from this study are as the following. First, lessons with the application of drawing tasks were effective in changing conception among the gifted science students. It is possibly because in the process where one student compare his/her own drawings with the others' ones and discuss them, changes in conception occurred effectively among the learners. Second, it was revealed that lessons utilizing drawing tasks have equal effects on changes in conception among both student groups irrespective of their levels of meta-cognition. Accordingly the lesson for changing perceptions utilizing drawing tasks developed in this study is a program which can be applied to all gifted science students in order to change conception among them. Third, lessons utilizing drawing tasks have the greatest effects on the gifted science students at a 'middle' level of creative problem solving abilities. Fourth, lessons utilizing drawing tasks have the greatest effects on the gifted science students at a 'middle' level of scientific inquiry skills. Putting these results together, it is thought that if lessons utilizing drawing tasks are applied to gifted science students, not only their concepts would be changed effectively but also their attitudes toward science would be changed positively.