Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Eun-Sung Jung has a total of 60 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Toward Acceleration of Variational Quantum Classifier Simulation on GPUs

    by Lan Anh Nguyen, Thai Anh Vo, Trang Nguyen, Son-Trung Doan, Toan-Duc Nguyen, Son-Hong Ngo, Eun-Sung Jung

  • Open Access

    Toward Generating Quality Test Questions and Answers Using Quantized Low-Rank Adapters in LLMs

    by Jebum Choi, Seongjun Hong, Seoyoon Hong, Jiyeon Park, Eun-Sung Jung.

    Traditional approaches to question-and-answer generation are resource-intensive, necessitating innovative automation techniques to address these challenges. To this end, we propose fine-tuning strategies based on Quantized Low-Rank Adaptation (QLoRA), utilizing a meticulously curated domain-specific dataset. As a case study, we focused on the Korea College Scholastic Ability Test (KCSAT), introducing the KCSAT-ENG dataset, which comprises questions and answers from real and mock tests. Our approach involved fine-tuning the LLaMA-3-8B-Instruct model to generate questions and answers for 22 distinct task types. A key innovation of our work is the deployment of separate models for question-and-answer generation, leveraging a cross-verification process to enhance accuracy. To evaluate the QLoRA technique, we conducted extensive experiments by varying quantization, rank, and alpha values. The results highlighted optimal configurations: the question generation model performed best with rank, α = 32, 8 respectively without quantization, while the answer generation model achieved optimal results with rank, α = 64, 16 respectively. Compared to a non-fine-tuned LLaMA-3-8B-Instruct model, our question generation model demonstrated a 41.5% improvement, and the answer generation model achieved a 16.1% improvement. These findings underscore the potential of QLoRA-based fine-tuning in creating accurate, cost-effective, and scalable automated educational tools.


  • Open Access

    Design and Implementation of an Automated Disaster-Recovery System for a Kubernetes Cluster Using LSTM

    by Ji-Beom Kim, Je-Bum Choi, Eun-Sung Jung

    With the increasing importance of data in modern business environments, effective data management and protection strategies are gaining increasing research attention. Data protection in a cloud environment is crucial for safeguarding information assets and maintaining sustainable services. This study introduces a system structure that integrates Kubernetes management platforms with backup and restoration tools. This system is designed to immediately detect disasters and automatically recover applications from another Kubernetes cluster. The experimental results show that this system executes the restoration process within 15 s without human intervention, enabling rapid recovery. This, in turn, significantly reduces the potential for delays and errors compared to manual recovery processes, thereby enhancing data management and recovery efficiency in cloud environments. Moreover, our research model predicts the CPU utilization of the cluster using Long Short-Term Memory (LSTM). The necessity of scheduling through this predict is made clearer through comparison with experiments without scheduling, demonstrating its ability to prevent performance degradation. This research highlights the efficiency and necessity of automatic recovery systems in cloud environments, setting a new direction for future research.


  • Open Access

    Design and Implementation of an Automated Disaster-recovery System for a Kubernetes Cluster Using LSTM

    by Ji-Beom Kim, Je-Bum Choi, Eun-Sung Jung

    With the increasing importance of data in the modern business environment, effective data man-agement and protection strategies are gaining increasing research attention. Data protection in a cloud environment is crucial for safeguarding information assets and maintaining sustainable services. This study introduces a system structure that integrates Kubernetes management plat-forms with backup and restoration tools. This system is designed to immediately detect disasters and automatically recover applications from another kubernetes cluster. The experimental results show that this system executes the restoration process within 15 s without human intervention, enabling rapid recovery. This, in turn, significantly reduces the potential for delays and errors compared with manual recovery processes, thereby enhancing data management and recovery ef-ficiency in cloud environments. Moreover, our research model predicts the CPU utilization of the cluster using Long Short-Term Memory (LSTM). The necessity of scheduling through this predict is made clearer through comparison with experiments without scheduling, demonstrating its ability to prevent performance degradation. This research highlights the efficiency and necessity of automatic recovery systems in cloud environments, setting a new direction for future research.

  • Open Access

    RLECN—A learning based dynamic threshold control of ECN

    by Shahzad, Eun-Sung Jung, Hyung Seok Kim


  • Open Access

    Optimizing Numerical Weather Prediction Model Performance Using Machine Learning Techniques

    by Soohyuck Choi, Eun-Sung Jung

    Weather forecasting primarily uses numerical weather prediction models that use weather observation data, including temperature and humidity, to predict future weather. The Korea Meteorological Administration (KMA) has adopted the GloSea6 numerical weather prediction model from the UK for weather forecasting. Besides utilizing these models for real-time weather forecasts, supercomputers are essential for running them for research purposes. However, owing to the limited supercomputer resources, many researchers have faced difficulties running the models. To address this issue, the KMA has developed a low-resolution model called Low GloSea6, which can be run on small and medium-sized servers in research institutions, but Low GloSea6 still uses numerous computer resources, especially in the I/O load. As I/O load can cause performance degradation for models with high data I/O, model I/O optimization is essential, but trial-and-error optimization by users is inefficient. Therefore, this study presents a machine learning-based approach to optimize the hardware and software parameters of the Low GloSea6 research environment. The proposed method comprised two steps. First, performance data were collected using profiling tools to obtain hardware platform parameters and Low GloSea6 internal parameters under various settings. Second, a machine learning model was trained using the collected data to determine the optimal hardware platform parameters and Low GloSea6 internal parameters for new research environments. The machine-learning model successfully predicted the optimal parameter combinations in different research environments, exhibiting a high degree of accuracy compared to the actual parameter combinations. In particular, the predicted model execution time based on the parameter combination showed a significant outcome with an error rate of only 16% compared to the actual execution time. Overall, this optimization method holds the potential to improve the performance of other high-performance computing scientific applications.


  • Open Access

    Comparative Study on Distributed Lightweight Deep Learning Models for Road Pothole Detection

    by Hassam Tahir, Eun-Sung Jung

    This paper delves into image detection based on distributed deep-learning techniques for intelligent traffic systems or self-driving cars. The accuracy and precision of neural networks deployed on edge devices (e.g., CCTV (closed-circuit television) for road surveillance) with small datasets may be compromised, leading to the misjudgment of targets. To address this challenge, TensorFlow and PyTorch were used to initialize various distributed model parallel and data parallel techniques. Despite the success of these techniques, communication constraints were observed along with certain speed issues. As a result, a hybrid pipeline was proposed, combining both dataset and model distribution through an all-reduced algorithm and NVlinks to prevent miscommunication among gradients. The proposed approach was tested on both an edge cluster and Google cluster environment, demonstrating superior performance compared to other test settings, with the quality of the bounding box detection system meeting expectations with increased reliability. Performance metrics, including total training time, images/second, cross-entropy loss, and total loss against the number of the epoch, were evaluated, revealing a robust competition between TensorFlow and PyTorch. The PyTorch environment’s hybrid pipeline outperformed other test settings.


  • Open Access

    Improving Air Pollution Prediction System through Multimodal Deep Learning Model Optimization

    by Kyung-Kyu Ko, Eun-Sung Jung

    Many forms of air pollution increase as science and technology rapidly advance. In particular, fine dust harms the human body, causing or worsening heart and lung-related diseases. In this study, the level of fine dust in Seoul after 8 h is predicted to prevent health damage in advance. We construct a dataset by combining two modalities (i.e., numerical and image data) for accurate prediction. In addition, we propose a multimodal deep learning model combining a Long Short Term Memory (LSTM) and Convolutional Neural Network (CNN). An LSTM AutoEncoder is chosen as a model for numerical time series data processing and basic CNN. A Visual Geometry Group Neural Network (VGGNet) (VGG16, VGG19) is also chosen as a CNN model for image processing to compare performance differences according to network depth. The VGGNet is a standard deep CNN architecture with multiple layers. Our multimodal deep learning model using two modalities (i.e., numerical and image data) showed better performance than a single deep learning model using only one modality (numerical data). Specifically, the performance improved up to 14.16% when the VGG19 model, which has a deeper network, was used rather than the VGG16 model.


  • Open Access

    Development of Cybersecurity Technology and Algorithm Based on Quantum Computing

    by Kyung-Kyu Ko, Eun-Sung Jung

    Many hacking incidents are linked to work files because most companies work with them. However, a variety of file encryption and decryption methods have been proposed. Existing file encryption/decryption technologies are under threat as hacking technologies advance, necessitating the development of stronger encryption algorithms. Therefore, in this study, we propose a modified advanced encryption standard (AES) algorithm and use quantum computing to encrypt/decrypt AES image files. Because the shift is regular during the AES Shift Row procedure, the change technique led the shift to become irregular when using quantum random walk. Computing resources and speeds were simulated using IBM Qiskit quantum simulators for performance evaluation, whereas encryption performance was assessed using number of pixels change rate (NPCR) and unified average changing intensity (UACI).