Jeong-Joon Kim
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
Jeong-Joon Kim has a total of 11 co-authors. Here are the top co-authors by number of shared publications:
- Dong-Jin Shin
Department of Computer Engineering, Anyang University, Anyang-si 14058, Korea
4 shared publications Jin-Yong Moon
Major in Online Video Promotion Gangdong University 278 Daehak-gil Gamgok-myeon Eumseong-gun Chungcheongbuk-do 27600 Korea
4 shared publicationsSeung-Yeon Hwang
Anyang Univ, Dept Comp Engn, Anyang Si 14058, South Korea
3 shared publications- Dong-Oh Kim
Electronics and Telecommunications Research Institute
2 shared publications - Hong-Yeon Kim
Electronics and Telecommunications Research Institute
2 shared publications - Jeongmin Park
Department of Computer Engineering, Tech University of Korea, Sangidaehak-ro, Siheung-si, Republic of South Korea
2 shared publications
Latest Publications
- Open Access
- 30 December 2025
- Published by Academic Society for Appropriate Technology in Journal of Appropriate Technology
- Vol. 11 (3) , 191-198
Web-based Learning Schedule Management System
Studying for self-development is a lifelong necessity, regardless of age, for students, office workers, and individuals in all walks of life. Recently, trends such as measuring personal study time with a stopwatch and sharing it on social media, or participating in live “Study With Me” sessions on YouTube, have gained popularity. These practices help reduce the sense of isolation in solo studying and foster motivation through shared experiences and a sense of community. In this paper, we developed a web-based platform that allows users to track their study time in real time and visualize the data through various statistics. This system helps individuals manage their study schedules more efficiently and encourages better time management. In addition, users can simulate the experience of studying with others through group study features, which adds a sense of companionship during study sessions. A daily ranking system based on total study time adds a competitive element that may motivate users to study longer and more consistently. Furthermore, the platform includes a bulletin board where users can exchange study tips, share useful resources, and discuss certification-related information. This creates an active community and enhances the overall learning experience through peer interaction. Future enhancements may include a mobile application and personalized feedback features tailored to specific study areas, aiming to meet a wider range of user needs and preferences.
- 30 December 2025
- Published by Academic Society for Appropriate Technology in Journal of Appropriate Technology
- Vol. 11 (3) , 191-198
Open AccessWeb-based Learning Schedule Management System
Studying for self-development is a lifelong necessity, regardless of age, for students, office workers, and individuals in all walks of life. Recently, trends such as measuring personal study time with a stopwatch and sharing it on social media, or participating in live “Study With Me” sessions on YouTube, have gained popularity. These practices help reduce the sense of isolation in solo studying and foster motivation through shared experiences and a sense of community. In this paper, we developed a web-based platform that allows users to track their study time in real time and visualize the data through various statistics. This system helps individuals manage their study schedules more efficiently and encourages better time management. In addition, users can simulate the experience of studying with others through group study features, which adds a sense of companionship during study sessions. A daily ranking system based on total study time adds a competitive element that may motivate users to study longer and more consistently. Furthermore, the platform includes a bulletin board where users can exchange study tips, share useful resources, and discuss certification-related information. This creates an active community and enhances the overall learning experience through peer interaction. Future enhancements may include a mobile application and personalized feedback features tailored to specific study areas, aiming to meet a wider range of user needs and preferences.
- Open Access
- 30 August 2025
- Published by Academic Society for Appropriate Technology in Journal of Appropriate Technology
- Vol. 11 (2) , 89-102
Development of a Product Purchase Assistance Application for the Visually Impaired
The purpose of this development is to support visually impaired individuals in receiving accurate product information during convenience store purchases, thereby ensuring their rights as consumers. Currently, visually impaired users face significant difficulties in identifying products due to insufficient braille labeling, especially on cylindrical items such as beverage cans, where the braille is often small, inconsistently placed, or even missing. This makes independent product selection challenging and often unreliable. To address these challenges, the system provides navigation to the nearest convenience store based on the user’s current location. Once inside the store, users can scan the front of a desired product using their smartphone camera. The system employs image recognition technology to analyze the product’s front image and extract key information such as product name, brand, and main ingredients. This information is then conveyed to the user through voice output, enabling them to make informed purchasing decisions independently. This application goes beyond simply enhancing accessibility; it empowers visually impaired users to engage in consumer activities without assistance. Additionally, the solution offers potential for future expansion, including support for multiple product categories, multilingual functionality, and features like real-time pricing and discount alerts. Ultimately, it lays the foundation for a more inclusive and intelligent retail assistance system.
- 30 August 2025
- Published by Academic Society for Appropriate Technology in Journal of Appropriate Technology
- Vol. 11 (2) , 89-102
Open AccessDevelopment of a Product Purchase Assistance Application for the Visually Impaired
The purpose of this development is to support visually impaired individuals in receiving accurate product information during convenience store purchases, thereby ensuring their rights as consumers. Currently, visually impaired users face significant difficulties in identifying products due to insufficient braille labeling, especially on cylindrical items such as beverage cans, where the braille is often small, inconsistently placed, or even missing. This makes independent product selection challenging and often unreliable. To address these challenges, the system provides navigation to the nearest convenience store based on the user’s current location. Once inside the store, users can scan the front of a desired product using their smartphone camera. The system employs image recognition technology to analyze the product’s front image and extract key information such as product name, brand, and main ingredients. This information is then conveyed to the user through voice output, enabling them to make informed purchasing decisions independently. This application goes beyond simply enhancing accessibility; it empowers visually impaired users to engage in consumer activities without assistance. Additionally, the solution offers potential for future expansion, including support for multiple product categories, multilingual functionality, and features like real-time pricing and discount alerts. Ultimately, it lays the foundation for a more inclusive and intelligent retail assistance system.
- Open Access
- 30 August 2025
- Published by Academic Society for Appropriate Technology in Journal of Appropriate Technology
- Vol. 11 (2) , 74-88
Online Learning Support Systems
After the COVID-19 pandemic, the education system rapidly shifted to cyber-based learning. As a result, there was a sharp increase in demand for online lectures, but the issue of lacking study materials remained unresolved. In particular, video-only lectures make it difficult for learners to organize and review content effectively. To address this issue, this study proposes a system that automatically extracts PDF materials from lecture videos. The extracted materials are processed using OCR technology, allowing users to convert them into text and add their own notes. Additionally, users can store and repeatedly memorize key terms, facilitating more efficient learning. A calendar feature is available to help users plan and manage their study schedules. Moreover, a sharing system enables collaborative learning by allowing users to exchange study materials. This system is designed to enhance the effectiveness of online learning and promote self-directed study. Therefore, this research provides a crucial solution for improving the quality of online learning environments.
- 30 August 2025
- Published by Academic Society for Appropriate Technology in Journal of Appropriate Technology
- Vol. 11 (2) , 74-88
Open AccessOnline Learning Support Systems
After the COVID-19 pandemic, the education system rapidly shifted to cyber-based learning. As a result, there was a sharp increase in demand for online lectures, but the issue of lacking study materials remained unresolved. In particular, video-only lectures make it difficult for learners to organize and review content effectively. To address this issue, this study proposes a system that automatically extracts PDF materials from lecture videos. The extracted materials are processed using OCR technology, allowing users to convert them into text and add their own notes. Additionally, users can store and repeatedly memorize key terms, facilitating more efficient learning. A calendar feature is available to help users plan and manage their study schedules. Moreover, a sharing system enables collaborative learning by allowing users to exchange study materials. This system is designed to enhance the effectiveness of online learning and promote self-directed study. Therefore, this research provides a crucial solution for improving the quality of online learning environments.
- Open Access
- 30 April 2025
- Published by Academic Society for Appropriate Technology in Journal of Appropriate Technology
- Vol. 11 (1) , 1-7
3D Smart Farm Simulation
In modern agricultural societies, smart farm systems that allow for both autonomous and manual adjustment of farming environments are increasingly being adopted. These systems contribute to improved productivity by enabling real-time monitoring of crop conditions and reducing the labor burden on farmers through automated control. Consequently, many farmers from various socioeconomic backgrounds have begun implementing smart farming technologies. However, a growing number of these users find it difficult to fully experience the expected convenience and efficiency, and due to the high initial installation costs and disappointing outcomes, many farmers are withdrawing their facilities. For smart farm projects to succeed and be widely accepted, it is essential not only to provide the technology but also to enable farmers to understand and experience its efficiency firsthand. This study proposes a simulation system designed to allow users to manually adjust the types of smart farm facilities installed based on varying levels of investment budgets. The simulation offers a virtual environment in which users can indirectly experience the functions of a smart farm system, and it provides tools to analyze and compare the efficiency and impact of each component quantitatively. Through this experience, members of agricultural communities will gain a deeper understanding of the practical value and necessity of smart farming technologies. Ultimately, this approach aims to reduce the barriers to adoption, promote the sustainable implementation of smart farming solutions, and support the development of a more efficient and resilient agricultural infrastructure.
- 30 April 2025
- Published by Academic Society for Appropriate Technology in Journal of Appropriate Technology
- Vol. 11 (1) , 1-7
Open Access3D Smart Farm Simulation
In modern agricultural societies, smart farm systems that allow for both autonomous and manual adjustment of farming environments are increasingly being adopted. These systems contribute to improved productivity by enabling real-time monitoring of crop conditions and reducing the labor burden on farmers through automated control. Consequently, many farmers from various socioeconomic backgrounds have begun implementing smart farming technologies. However, a growing number of these users find it difficult to fully experience the expected convenience and efficiency, and due to the high initial installation costs and disappointing outcomes, many farmers are withdrawing their facilities. For smart farm projects to succeed and be widely accepted, it is essential not only to provide the technology but also to enable farmers to understand and experience its efficiency firsthand. This study proposes a simulation system designed to allow users to manually adjust the types of smart farm facilities installed based on varying levels of investment budgets. The simulation offers a virtual environment in which users can indirectly experience the functions of a smart farm system, and it provides tools to analyze and compare the efficiency and impact of each component quantitatively. Through this experience, members of agricultural communities will gain a deeper understanding of the practical value and necessity of smart farming technologies. Ultimately, this approach aims to reduce the barriers to adoption, promote the sustainable implementation of smart farming solutions, and support the development of a more efficient and resilient agricultural infrastructure.
- Open Access
- 30 April 2025
- Published by Academic Society for Appropriate Technology in Journal of Appropriate Technology
- Vol. 11 (1) , 8-15
Development of Auxiliary Device Services for Shared Bicycle Business
In recent years, the rapid rise of the sharing economy has led to a proliferation of rental-based services in various industries. A notable example is the widespread adoption of shared bicycles, with services like ‘Ddareungi’ and ‘Kakao T Bike’ now commonly available in most localities, often with at least two to three options per region. The key advantage of the sharing economy lies in its shift from ownership to access, promoting the efficient utilization of resources and fostering a collaborative consumption model aimed at sustainability. However, this sector now faces intense market competition and persistent challenges related to profitability and differentiation. To address these issues, this study proposes a service enhancement model that integrates a Bluetooth Low Energy (BLE)-based auxiliary device as an optional feature for shared rental items. By offering value-added services that incur additional charges, the proposed system seeks to improve profitability while simultaneously enhancing user satisfaction and safety. This approach aims to transcend traditional rental models by offering enriched user experiences, thereby contributing to the structural advancement and sustainability of the sharing economy.
- 30 April 2025
- Published by Academic Society for Appropriate Technology in Journal of Appropriate Technology
- Vol. 11 (1) , 8-15
Open AccessDevelopment of Auxiliary Device Services for Shared Bicycle Business
In recent years, the rapid rise of the sharing economy has led to a proliferation of rental-based services in various industries. A notable example is the widespread adoption of shared bicycles, with services like ‘Ddareungi’ and ‘Kakao T Bike’ now commonly available in most localities, often with at least two to three options per region. The key advantage of the sharing economy lies in its shift from ownership to access, promoting the efficient utilization of resources and fostering a collaborative consumption model aimed at sustainability. However, this sector now faces intense market competition and persistent challenges related to profitability and differentiation. To address these issues, this study proposes a service enhancement model that integrates a Bluetooth Low Energy (BLE)-based auxiliary device as an optional feature for shared rental items. By offering value-added services that incur additional charges, the proposed system seeks to improve profitability while simultaneously enhancing user satisfaction and safety. This approach aims to transcend traditional rental models by offering enriched user experiences, thereby contributing to the structural advancement and sustainability of the sharing economy.
- Open Access
Cache-Based Matrix Technology for Efficient Write and Recovery in Erasure Coding Distributed File Systems
With the development of various information and communication technologies, the amount of big data has increased, and distributed file systems have emerged to store them stably. The replication technique divides the original data into blocks and writes them on multiple servers for redundancy and fault tolerance. However, there is a symmetrical space efficiency problem that arises from the need to store blocks larger than the original data. When storing data, the Erasure Coding (EC) technique generates parity blocks through encoding calculations and writes them separately on each server for fault tolerance and data recovery purposes. Even if a specific server fails, original data can still be recovered through decoding calculations using the parity blocks stored on the remaining servers. However, matrices generated during encoding and decoding are redundantly generated during data writing and recovery, which leads to unnecessary overhead in distributed file systems. This paper proposes a cache-based matrix technique that uploads the matrices generated during encoding and decoding to cache memory and reuses them, rather than generating new matrices each time encoding or decoding occurs. The design of the cache memory applies the Weighting Size and Cost Replacement Policy (WSCRP) algorithm to efficiently upload and reuse matrices to cache memory using parameters known as weights and costs. Furthermore, the cache memory table can be managed efficiently because the weight–cost model sorts and updates matrices using specific parameters, which reduces replacement cost. The experiment utilized the Hadoop Distributed File System (HDFS) as the distributed file system, and the EC volume was composed of Reed–Solomon code with parameters (6, 3). As a result of the experiment, it was possible to reduce the write, read, and recovery times associated with encoding and decoding. In particular, for up to three node failures, systems using WSCRP were able to reduce recovery time by about 30 s compared to regular HDFS systems.
Open AccessCache-Based Matrix Technology for Efficient Write and Recovery in Erasure Coding Distributed File Systems
With the development of various information and communication technologies, the amount of big data has increased, and distributed file systems have emerged to store them stably. The replication technique divides the original data into blocks and writes them on multiple servers for redundancy and fault tolerance. However, there is a symmetrical space efficiency problem that arises from the need to store blocks larger than the original data. When storing data, the Erasure Coding (EC) technique generates parity blocks through encoding calculations and writes them separately on each server for fault tolerance and data recovery purposes. Even if a specific server fails, original data can still be recovered through decoding calculations using the parity blocks stored on the remaining servers. However, matrices generated during encoding and decoding are redundantly generated during data writing and recovery, which leads to unnecessary overhead in distributed file systems. This paper proposes a cache-based matrix technique that uploads the matrices generated during encoding and decoding to cache memory and reuses them, rather than generating new matrices each time encoding or decoding occurs. The design of the cache memory applies the Weighting Size and Cost Replacement Policy (WSCRP) algorithm to efficiently upload and reuse matrices to cache memory using parameters known as weights and costs. Furthermore, the cache memory table can be managed efficiently because the weight–cost model sorts and updates matrices using specific parameters, which reduces replacement cost. The experiment utilized the Hadoop Distributed File System (HDFS) as the distributed file system, and the EC volume was composed of Reed–Solomon code with parameters (6, 3). As a result of the experiment, it was possible to reduce the write, read, and recovery times associated with encoding and decoding. In particular, for up to three node failures, systems using WSCRP were able to reduce recovery time by about 30 s compared to regular HDFS systems.
- Open Access
- 1 January 2023
- Published by Tech Science Press in Computers, Materials & Continua
- Vol. 75 (3) , 4669-4680
Data Augmentation Using Contour Image for Convolutional Neural Network
With the development of artificial intelligence-related technologies such as deep learning, various organizations, including the government, are making various efforts to generate and manage big data for use in artificial intelligence. However, it is difficult to acquire big data due to various social problems and restrictions such as personal information leakage. There are many problems in introducing technology in fields that do not have enough training data necessary to apply deep learning technology. Therefore, this study proposes a mixed contour data augmentation technique, which is a data augmentation technique using contour images, to solve a problem caused by a lack of data. ResNet, a famous convolutional neural network (CNN) architecture, and CIFAR-10, a benchmark data set, are used for experimental performance evaluation to prove the superiority of the proposed method. And to prove that high performance improvement can be achieved even with a small training dataset, the ratio of the training dataset was divided into 70%, 50%, and 30% for comparative analysis. As a result of applying the mixed contour data augmentation technique, it was possible to achieve a classification accuracy improvement of up to 4.64% and high accuracy even with a small amount of data set. In addition, it is expected that the mixed contour data augmentation technique can be applied in various fields by proving the excellence of the proposed data augmentation technique using benchmark datasets.
- 1 January 2023
- Published by Tech Science Press in Computers, Materials & Continua
- Vol. 75 (3) , 4669-4680
Open AccessData Augmentation Using Contour Image for Convolutional Neural Network
With the development of artificial intelligence-related technologies such as deep learning, various organizations, including the government, are making various efforts to generate and manage big data for use in artificial intelligence. However, it is difficult to acquire big data due to various social problems and restrictions such as personal information leakage. There are many problems in introducing technology in fields that do not have enough training data necessary to apply deep learning technology. Therefore, this study proposes a mixed contour data augmentation technique, which is a data augmentation technique using contour images, to solve a problem caused by a lack of data. ResNet, a famous convolutional neural network (CNN) architecture, and CIFAR-10, a benchmark data set, are used for experimental performance evaluation to prove the superiority of the proposed method. And to prove that high performance improvement can be achieved even with a small training dataset, the ratio of the training dataset was divided into 70%, 50%, and 30% for comparative analysis. As a result of applying the mixed contour data augmentation technique, it was possible to achieve a classification accuracy improvement of up to 4.64% and high accuracy even with a small amount of data set. In addition, it is expected that the mixed contour data augmentation technique can be applied in various fields by proving the excellence of the proposed data augmentation technique using benchmark datasets.
- Open Access
- 1 January 2023
- Published by Tech Science Press in Computers, Materials & Continua
- Vol. 75 (2) , 3553-3569
A Universal Activation Function for Deep Learning
Recently, deep learning has achieved remarkable results in fields that require human cognitive ability, learning ability, and reasoning ability. Activation functions are very important because they provide the ability of artificial neural networks to learn complex patterns through nonlinearity. Various activation functions are being studied to solve problems such as vanishing gradients and dying nodes that may occur in the deep learning process. However, it takes a lot of time and effort for researchers to use the existing activation function in their research. Therefore, in this paper, we propose a universal activation function (UA) so that researchers can easily create and apply various activation functions and improve the performance of neural networks. UA can generate new types of activation functions as well as functions like traditional activation functions by properly adjusting three hyperparameters. The famous Convolutional Neural Network (CNN) and benchmark dataset were used to evaluate the experimental performance of the UA proposed in this study. We compared the performance of the artificial neural network to which the traditional activation function is applied and the artificial neural network to which the UA is applied. In addition, we evaluated the performance of the new activation function generated by adjusting the hyperparameters of the UA. The experimental performance evaluation results showed that the classification performance of CNNs improved by up to 5% through the UA, although most of them showed similar performance to the traditional activation function.
- 1 January 2023
- Published by Tech Science Press in Computers, Materials & Continua
- Vol. 75 (2) , 3553-3569
Open AccessA Universal Activation Function for Deep Learning
Recently, deep learning has achieved remarkable results in fields that require human cognitive ability, learning ability, and reasoning ability. Activation functions are very important because they provide the ability of artificial neural networks to learn complex patterns through nonlinearity. Various activation functions are being studied to solve problems such as vanishing gradients and dying nodes that may occur in the deep learning process. However, it takes a lot of time and effort for researchers to use the existing activation function in their research. Therefore, in this paper, we propose a universal activation function (UA) so that researchers can easily create and apply various activation functions and improve the performance of neural networks. UA can generate new types of activation functions as well as functions like traditional activation functions by properly adjusting three hyperparameters. The famous Convolutional Neural Network (CNN) and benchmark dataset were used to evaluate the experimental performance of the UA proposed in this study. We compared the performance of the artificial neural network to which the traditional activation function is applied and the artificial neural network to which the UA is applied. In addition, we evaluated the performance of the new activation function generated by adjusting the hyperparameters of the UA. The experimental performance evaluation results showed that the classification performance of CNNs improved by up to 5% through the UA, although most of them showed similar performance to the traditional activation function.
- Open Access
- 7 April 2022
- Published by MDPI AG in Applied Sciences
- Vol. 12 (8) , 3734
A Deep Learning Framework Performance Evaluation to Use YOLO in Nvidia Jetson Platform
Deep learning-based object detection technology can efficiently infer results by utilizing graphics processing units (GPU). However, when using general deep learning frameworks in embedded systems and mobile devices, processing functionality is limited. This allows deep learning frameworks such as TensorFlow-Lite (TF-Lite) and TensorRT (TRT) to be optimized for different hardware. Therefore, this paper introduces a performance inference method that fuses the Jetson monitoring tool with TensorFlow and TRT source code on the Nvidia Jetson AGX Xavier platform. In addition, central processing unit (CPU) utilization, GPU utilization, object accuracy, latency, and power consumption of the deep learning framework were compared and analyzed. The model is You Look Only Once Version4 (YOLOv4), and the dataset uses Common Objects in Context (COCO) and PASCAL Visual Object Classes (VOC). We confirmed that using TensorFlow results in high latency. We also confirmed that TensorFlow-TensorRT (TF-TRT) and TRT using Tensor Cores provide the most efficiency. However, it was confirmed that TF-Lite showed the lowest performance because it utilizes a GPU limited to mobile devices. Through this paper, we think that when developing deep learning-related object detection technology on the Nvidia Jetson platform or desktop environment, services and research can be efficiently conducted through measurement results.
- 7 April 2022
- Published by MDPI AG in Applied Sciences
- Vol. 12 (8) , 3734
Open AccessA Deep Learning Framework Performance Evaluation to Use YOLO in Nvidia Jetson Platform
Deep learning-based object detection technology can efficiently infer results by utilizing graphics processing units (GPU). However, when using general deep learning frameworks in embedded systems and mobile devices, processing functionality is limited. This allows deep learning frameworks such as TensorFlow-Lite (TF-Lite) and TensorRT (TRT) to be optimized for different hardware. Therefore, this paper introduces a performance inference method that fuses the Jetson monitoring tool with TensorFlow and TRT source code on the Nvidia Jetson AGX Xavier platform. In addition, central processing unit (CPU) utilization, GPU utilization, object accuracy, latency, and power consumption of the deep learning framework were compared and analyzed. The model is You Look Only Once Version4 (YOLOv4), and the dataset uses Common Objects in Context (COCO) and PASCAL Visual Object Classes (VOC). We confirmed that using TensorFlow results in high latency. We also confirmed that TensorFlow-TensorRT (TF-TRT) and TRT using Tensor Cores provide the most efficiency. However, it was confirmed that TF-Lite showed the lowest performance because it utilizes a GPU limited to mobile devices. Through this paper, we think that when developing deep learning-related object detection technology on the Nvidia Jetson platform or desktop environment, services and research can be efficiently conducted through measurement results.
- Open Access
Systematic Review on Identification and Prediction of Deep Learning-Based Cyber Security Technology and Convergence Fields
Recently, as core technologies leading the fourth industrial revolution, such as the Internet of Things (IoT), 5G, the cloud, and big data, have promoted smart convergence across national socio-economic infrastructures, cyber systems are expanding and becoming complex, and they are not effective in responding to cyber safety risks and threats using security technology solutions limited to a single system. Therefore, we developed cyber security technology that combines machine learning and AI technology to solve complex problems related to cyber safety. In this regard, this study aims to identify technology development trends to prevent the risks and threats of various cyber systems by monitoring major cyber security convergence fields and technologies through the symmetrical thesis and patent analysis. Because thesis information can explain the superiority of technology and patent information can explain the usefulness of a technology, they can be effectively used for analyzing and predicting technology development trends. Therefore, in this study, latent Dirichlet allocation is applied to extract text-document-based technical topics for the symmetrical thesis and patent information to identify security convergence fields and technologies for cyber safety. In addition, it elucidates cyber security convergence fields and technology trends by applying a dynamic topic model and long short-term memory, which are useful for analyzing technological changes and predicting trends. Based on these results, cyber security administrators, system operators, and developers can effectively identify and respond to trends in related technologies to reduce threats, and companies and experts developing cyber security solutions can present a new security approach.
Open AccessSystematic Review on Identification and Prediction of Deep Learning-Based Cyber Security Technology and Convergence Fields
Recently, as core technologies leading the fourth industrial revolution, such as the Internet of Things (IoT), 5G, the cloud, and big data, have promoted smart convergence across national socio-economic infrastructures, cyber systems are expanding and becoming complex, and they are not effective in responding to cyber safety risks and threats using security technology solutions limited to a single system. Therefore, we developed cyber security technology that combines machine learning and AI technology to solve complex problems related to cyber safety. In this regard, this study aims to identify technology development trends to prevent the risks and threats of various cyber systems by monitoring major cyber security convergence fields and technologies through the symmetrical thesis and patent analysis. Because thesis information can explain the superiority of technology and patent information can explain the usefulness of a technology, they can be effectively used for analyzing and predicting technology development trends. Therefore, in this study, latent Dirichlet allocation is applied to extract text-document-based technical topics for the symmetrical thesis and patent information to identify security convergence fields and technologies for cyber safety. In addition, it elucidates cyber security convergence fields and technology trends by applying a dynamic topic model and long short-term memory, which are useful for analyzing technological changes and predicting trends. Based on these results, cyber security administrators, system operators, and developers can effectively identify and respond to trends in related technologies to reduce threats, and companies and experts developing cyber security solutions can present a new security approach.