• Journal of the Korea Society of Computer and Information
• /
• v.28 no.8
• /
• pp.195-204
• /
• 2023

In this study, prompt engineering that improves prompts was explored to improve the performance of teaching and learning materials recommendations using generative artificial intelligence such as GPT and Stable Diffusion. Picture materials were used as the types of teaching and learning materials. To explore the impact of the prompt composition, a Zero-Shot prompt, a prompt containing learning target grade information, a prompt containing learning goals, and a prompt containing both learning target grades and learning goals were designed to collect responses. The collected responses were embedded using Sentence Transformers, dimensionalized to t-SNE, and visualized, and then the relationship between prompts and responses was explored. In addition, each response was clustered using the k-means clustering algorithm, then the adjacent value of the widest cluster was selected as a representative value, imaged using Stable Diffusion, and evaluated by 30 elementary school teachers according to the criteria for evaluating teaching and learning materials. Thirty teachers judged that three of the four picture materials recommended were of educational value, and two of them could be used for actual classes. The prompt that recommended the most valuable picture material appeared as a prompt containing both the target grade and the learning goal.

• Journal of KIISE:Software and Applications
• /
• v.36 no.7
• /
• pp.523-530
• /
• 2009

Despite the growing need for customized operating system kernels for embedded devices, kernel development continues to suffer from insufficient reliability and high testing cost for several reasons such as the high complexity of the kernel code. To alleviate these difficulties, this study proposes the MOdel-based KERnel Testing (MOKERT) framework for detection of concurrency bugs in the kernel. MOKERT translates a given C program into a corresponding Promela model, and then tries to find a counter example with regard to a given requirement property, If found, MOKERT executes that counter example on the real kernel code to check whether the counter example is a false alarm or not, The MOKERT framework was applied to the Linux proc file system and confirmed that the bug reported in a ChangeLog actually caused a data race problem, In addition, a new data race bug in the Linux proc file system was found, which causes kernel panic.

• Journal of the Korean Society of Radiology
• /
• v.18 no.6
• /
• pp.715-724
• /
• 2024

We aimed to build an MR-CT interconversion model based on structure-constraints Cycle-constraints Generative Adversarial Neural Networks (scCycleGANs). We used MDCT (Somatom Definition Flash CT, SIEMENS, Germany) and 3.0T MRI (Ingenia 3.0T CX MRI, PHILIPS, Netherlands) as our hardware equipment and Python (3.12.6) and PyTorch (2.4.0) as software. The study model was scCycleGAN. We acquired 2,871 and 2,436 brain CT and MR (T2WI) images of 87 patients, respectively, and for a total of 5,307 medical images, CT and MR images taken at the same level were classified through primary evaluation, and 364, 27, and 8 pairs of images were labeled as training, validation, and test data, respectively. Then, we applied hybrid objective function to the GAN model based on the basic APS frameworks to build the model, and the evaluation of the generated model was divided into quantitative and qualitative evaluation. The qualitative evaluation was conducted on 10 radiologists with more than 20 years of experience, and the quantitative evaluation was set as PSNR, IOU, SSIM, and MAE. The results of the qualitative evaluation showed that the percentage of 'positive responses', defined as a response of 'Neutral' or better, was 63% and 96% for the Synthesis CT (sCT) and Synthesis MR (sMR) groups, respectively, while the quantitative evaluation metrics PSNR, SSIM, and MAE achieved the initial target values for both groups. Our study can be used as basic guided research in the field of medical image conversion and synthesis. And further research and complementary studies are expected to solve problems such as model lightweighting to reduce the dose burden on patients and medical costs if applied to clinical environments.

• Journal of the Korea Computer Graphics Society
• /
• v.20 no.2
• /
• pp.1-11
• /
• 2014

The quality of 2D-to-3D conversion depends on the accuracy of the assigned depth to scene objects. Manual depth painting for given objects is labor intensive as each frame is painted. Specifically, a human is one of the most challenging objects for a high-quality conversion, as a human body is an articulated figure and has many degrees of freedom (DOF). In addition, various styles of clothes, accessories, and hair create a very complex silhouette around the 2D human object. We propose an efficient method to estimate visually pleasing depths of a human at every frame in a monocular video. First, a 3D template model is matched to a person in a monocular video with a small number of specified user correspondences. Our pose estimation with sequential joint angular constraints reproduces a various range of human motions (i.e., spine bending) by allowing the utilization of a fully skinned 3D model with a large number of joints and DOFs. The initial depth of the 2D object in the video is assigned from the matched results, and then propagated toward areas where the depth is missing to produce a complete depth map. For the effective handling of the complex silhouettes and appearances, we introduce a partial depth propagation method based on color segmentation to ensure the detail of the results. We compared the result and depth maps painted by experienced artists. The comparison shows that our method produces viable depth maps of humans in monocular videos efficiently.

• Journal of Christian Education in Korea
• /
• v.78
• /
• pp.57-85
• /
• 2024

Purpose of the study: In order to utilize artificial intelligence-generated AI in educational ministry, this study analyzes the concept of artificial intelligence and generative AI and the educational theological aspects of educational ministry to find ways to efficiently utilize artificial intelligence ChatGPT in educational ministry. Contents and methods of the study: The contents of this study are. First, the contents of this study were analyzed by dividing the concepts of artificial intelligence and generative AI into the concept of artificial intelligence, types of artificial intelligence, and generative language model AI ChatGPT. Second, the educational theological analysis of educational ministry was divided into the concept of educational ministry, the goals of educational ministry, the content of educational ministry, and the direction of educational ministry in the era of artificial intelligence. Third, the plan to use artificial intelligence ChatGPT in educational ministry is to provide tools for writing sermon manuscripts, preparation tools for worship and prayer, and church education, focusing on the five functions of the early church community. It was analyzed by dividing it into tools for teaching, tools for teaching materials for believers, and tools for serving and volunteering. Conclusion and Recommendation: The conclusion of this study is that, first, when writing sermon manuscripts through artificial intelligence ChatGPT, high-quality sermon manuscripts can be written through the preacher's spirituality, faith, and insight. Second, through artificial intelligence ChatGPT, you can efficiently design and plan worship services and prepare services that serve the congregation objectively through various scenarios. Third, by using artificial intelligence ChatGPT in church education, it can be used while maintaining a complementary relationship with teachers through collaboration with human and artificial intelligence teachers. Fourth, through artificial intelligence ChatGPT, we provide a program that allows members of the church community to share spiritual fellowship, a plan to meet the needs of church members and strengthen interdependence, and an attitude of actively welcoming new people and respecting diversity. It provides useful materials that can play an important role in giving, loving, serving, and growing together in the love of Christ. Lastly, through artificial intelligence ChatGPT, we are seeking ways to provide various information about volunteer activities, learning support for children and youth in the community, mentoring-related programs, and playing a leading role in forming a village community in the local community.

• Journal of Digital Convergence
• /
• v.11 no.4
• /
• pp.251-258
• /
• 2013

Recently companies are confronted with a much more sophisticated business environment than before and at the same time have to be able to adapt to rapid changes. Accordingly, the need for selecting among alternatives and managing systematic decision-making has been steadily increasing to respond to a more diverse customer needs and keep up with the fierce competition. In this study, we propose a framework that consist of problem solving procedures and techniques and knowledge structure built on processes to support strategic decision making. and discuss how to utilize simulation tools as the knowledge-based problem solving tools. In addition we discuss how to build and advance the knowledge structure to implement the proposed architecture. Management decision support systems architecture consist of three key factors. The first is Problem Solving Approach which is used as reference. The second is knowledge structure on business processes that includes standard and reference business processes. The third is simulators that are able to generate and analyze alternatives using problem solving techniques and knowledge base. In sum, the proposed framework of decision-making support systems facilitates knowledge-intensive consulting processes to promote the development and application of consulting knowledge and techniques and increase the efficiency of consulting firms and industry.

• The Journal of the Acoustical Society of Korea
• /
• v.38 no.6
• /
• pp.703-709
• /
• 2019

This paper proposes speech recognition systems employing Convolutional Neural Network (CNN) and Long Short Term Memory (LSTM) structures combined with Hidden Markov Moldel (HMM) to effectively recognize the speech of VeloPharyngeal Insufficiency (VPI) patients, and compares the recognition performance of the systems to the Gaussian Mixture Model (GMM-HMM) and fully-connected Deep Neural Network (DNNHMM) based speech recognition systems. In this paper, the initial model is trained using normal speakers' speech and simulated VPI speech is used for generating a prior model for speaker adaptation. For VPI speaker adaptation, selected layers are trained in the CNN-HMM based model, and dropout regulatory technique is applied in the LSTM-HMM based model, showing 3.68 % improvement in recognition accuracy. The experimental results demonstrate that the proposed LSTM-HMM-based speech recognition system is effective for VPI speech with small-sized speech data, compared to conventional GMM-HMM and fully-connected DNN-HMM system.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.28 no.1
• /
• pp.184-192
• /
• 2022

Valve internal leakage is caused by damage to the internal parts of the valve, resulting in accidents and shutdowns of the piping system. This study investigated the possibility of a real-time leak detection method using the acoustic emission (AE) signal generated from the piping system during the internal leakage of a butterfly valve. Datasets of raw time-domain AE signals were collected and postprocessed for each operation mode of the valve in a systematic manner to develop a data-driven model for the detection and classification of internal leakage, by applying machine learning algorithms. The aim of this study was to determine whether it is possible to treat leak detection as a classification problem by applying two classification algorithms: support vector machine (SVM) and convolutional neural network (CNN). The results showed different performances for the algorithms and datasets used. The SVM-based binary classification models, based on feature extraction of data, achieved an overall accuracy of 83% to 90%, while in the case of a multiple classification model, the accuracy was reduced to 66%. By contrast, the CNN-based classification model achieved an accuracy of 99.85%, which is superior to those of any other models based on the SVM algorithm. The results revealed that the SVM classification model requires effective feature extraction of the AE signals to improve the accuracy of multi-class classification. Moreover, the CNN-based classification can be a promising approach to detect both leakage and valve opening as long as the performance of the processor does not degrade.

• Journal of the Korea Society of Computer and Information
• /
• v.14 no.2
• /
• pp.201-209
• /
• 2009

According to previous researches about online evaluation in many e-Learning contents, it took too much time and effort to generate test questions for formative or achievement tests using a database as an item pool. Furthermore, it is hard to measure accomplishment of learners for each unit through overall tests provided by existing e-learning contents. In this paper, to efficiently cope with problems described above, the item pool based on Item Form was transformed into Interaction Date Model in Run-Time Environment of SCORM2004. And the contents for the math concepts and principles that students would learn from regular classroom were developed in accordance with SCORM. In addition, Confidence Factor Function was used to take an objective view in measuring the accomplishment of learners through the items automatically generated by LMS(Learning Management System).

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.14 no.1
• /
• pp.72-78
• /
• 1990

An interactive computer program has been developed to design a pattern and risers for the production of castings of high quality. In our system, the user models the shape of a final product by using the system's modeling capability, a pattern is generated in a three dimensional model by eliminating the holes and adding shrinkage allowances and drafts, the proper riser is created automatically, and they are united together to yield the three dimensional model of the portion of a mold assembly. The mold can be completed after the runners and the gating systems are designed, modeled, and united, which will be described in part 2 of this work. The unique feature of this work is a realization of an automatic design of the pattern and risers by integrating the modeling capabilities and the design equations used in the real practice.