• Electronics and Telecommunications Trends
• /
• v.35 no.5
• /
• pp.1-13
• /
• 2020

Recently, network infrastructure has evolved into a BizTech agile autonomous network to cope with the dynamic changes in the service environment. This survey presents the expectations from two different perspectives of the harmonization of network and artificial intelligence (AI) technologies. First, the paper focuses on the possibilities of AI technology for the autonomous network industry. Subsequently, it discusses how networks can play a role in the evolution of distributed AI technologies.

• Electronics and Telecommunications Trends
• /
• v.35 no.7
• /
• pp.23-35
• /
• 2020

In this paper, we introduce ETRI AI Strategy #3, "Leading Future Technologies of Network, Media, and Content." Its first goal is "to innovate AI service technology to overcome the current limitations of AI technologies." Artificial intelligence (AI) services, such as self-driving cars and robots, are combinations of computing, network, AI algorithms, and other technologies. To develop AI services, we need to develop different types of network, media coding, and content creation technologies. Moreover, AI technologies are adopted in ICT technologies. Self-planning and self-managing networks and automatic content creation technologies using AI are being developed. This paper introduces the two directions of ETRI's ICT technology development plan for AI: ICT for AI and ICT by AI. The area of ICT for AI has only recently begun to develop. ETRI, the ICT leader, hopes to have opportunities for leadership in the second wave of AI services.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.35 no.12A
• /
• pp.1181-1187
• /
• 2010

This paper proposes a mobile AI (Artificial Intelligence) conducting decision-making in the game through education for intelligent character on the basis of Neural Network. Neural Network is learned through the input/output value of the algorithm which defines the game rule and the problem solving method. The learned character is able to perceive the circumstances and make proper action. In this paper, the mobile AI using Neural Network has been step-by-step designed, and a simple game has been materialized for its functional experiment. In this game, the goal, the character, and obstacles exist on regular 2D space, and the character, evading obstacles, has to move where the goal is. The mobile AI can achieve its goals in changing environment by learning the solution to several problems through the algorithm defined in each experiment. The defined algorithm and Neural Network are designed to make the input/output system the same. As the experimental results, the suggested mobile AI showed that it could perceive the circumstances to conduct action and to complete its mission. If mobile AI learns the defined algorithm even in the game of complex structure, its Neural Network will be able to show proper results even in the changing environment.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.33 no.3
• /
• pp.549-559
• /
• 2023

As mobile edge computing (MEC) is gaining attention as a core technology of 5G networks, edge AI technology of 5G network environment based on mobile user data is recently being used in various fields. However, as in traditional AI security, there is a possibility of adversarial interference of standard 5G network functions within the core network responsible for edge AI core functions. In addition, research on data poisoning attacks that can occur in the MEC environment of standalone mode defined in 5G standards by 3GPP is currently insufficient compared to existing LTE networks. In this study, we explore the threat model for the MEC environment using NWDAF, a network function that is responsible for the core function of edge AI in 5G, and propose a feature selection method to improve the performance of detecting data poisoning attacks for Leaf NWDAF as some proof of concept. Through the proposed methodology, we achieved a maximum detection rate of 94.9% for Slowloris attack-based data poisoning attacks in NWDAF.

• Journal of the Korean Society for information Management
• /
• v.41 no.3
• /
• pp.289-308
• /
• 2024

This study investigates the collaboration networks in the field of AI-driven diagnostic medical imaging, focusing on the influence of two social capital concepts-network closure and structural holes-on research performance. The analysis reveals a highly fragmented network structure with one dominant component, while individual clusters exhibit strong internal cohesion. Both network closure, measured by density, and structural holes, assessed through efficiency, positively impact research performance, as demonstrated by QAP regression analysis. The findings highlight that, in the integration of AI into diagnostic medical imaging, robust connections among researchers are vital, and the presence of structural holes, which enable the assimilation of diverse knowledge, also significantly enhances research outcomes. This underscores the importance of fostering a well-balanced network to optimize collaboration and knowledge production in this emerging interdisciplinary field.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.655-656
• /
• 2021

April 2021, Disaster Safety Communication Network services have been launched, but security functions are weak at the beginning of the service. The current security method for Android-based APP is using Google Protect's technology to detect malware. Malware is difficult to detect directly because there are various types, so by applying malware detection technology that combines AI and Google Protect technology to Disaster Safety Communication Networks, research on how to implement 'AI-based Disaster Satety Communication Network Protect'.

• Journal of the Institute of Convergence Signal Processing
• /
• v.23 no.4
• /
• pp.234-240
• /
• 2022

This paper proposes a customized AI exercise recommendation service for balancing the relative amount of exercise according to the working environment by each occupation. WISDM database is collected by using acceleration and gyro sensors, and is a dataset that classifies physical activities into 18 categories. Our system recommends a adaptive exercise using the analyzed activity type after classifying 18 physical activities into 3 physical activities types such as whole body, upper body and lower body. 1 Dimensional convolutional neural network is used for classifying a physical activity in this paper. Proposed model is composed of a convolution blocks in which 1D convolution layers with a various sized kernel are connected in parallel. Convolution blocks can extract a detailed local features of input pattern effectively that can be extracted from deep neural network models, as applying multi 1D convolution layers to input pattern. To evaluate performance of the proposed neural network model, as a result of comparing the previous recurrent neural network, our method showed a remarkable 98.4% accuracy.

• Journal of Information Technology Services
• /
• v.18 no.2
• /
• pp.123-141
• /
• 2019

The goal of this study is to review the major research trend on the convergence studies of AI and healthcare technologies. For the study, 15,260 English articles on AI and healthcare related topics were collected from Scopus for 55 years from 1963, and text mining techniques were conducted. As a result, seven key research topics were defined : "AI for Clinical Decision Support System (CDSS)", "AI for Medical Image", "Internet of Healthcare Things (IoHT)", "Big Data Analytics in Healthcare", "Medical Robotics", "Blockchain in Healthcare", and "Evidence Based Medicine (EBM)". The result of this study can be utilized to set up and develop the appropriate healthcare R&D strategies for the researchers and government. In this study, text mining techniques such as Text Analysis, Frequency Analysis, Topic Modeling on LDA (Latent Dirichlet Allocation), Word Cloud, and Ego Network Analysis were conducted.

• Journal of the Korea Convergence Society
• /
• v.13 no.1
• /
• pp.51-62
• /
• 2022

AI-based speed prediction studies have been conducted quite actively. However, while the importance of explainable AI is emerging, the study of interpreting and reasoning the AI-based speed predictions has not been carried out much. Therefore, in this paper, 'Explainable Deep Graph Neural Network (GNN)' is devised to analyze the speed prediction and assess the nearby road influence for reasoning the critical contributions to a given road situation. The model's output was explained by comparing the differences in output before and after masking the input values of the GNN model. Using TOPIS traffic speed data, we applied our GNN models for the major congested roads in Seoul. We verified our approach through a traffic flow simulation by adjusting the most influential nearby roads' speed and observing the congestion's relief on the road of interest accordingly. This is meaningful in that our approach can be applied to the transportation network and traffic flow can be improved by controlling specific nearby roads based on the inference results.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.27 no.2
• /
• pp.256-264
• /
• 2024

In the last 10 years, AI has made rapid progress, and image classification, in particular, are showing excellent performance based on deep learning. Nevertheless, due to the nature of deep learning represented by a black box, it is difficult to actually use it in critical decision-making situations such as national defense, autonomous driving, medical care, and finance due to the lack of explainability of judgement results. In order to overcome these limitations, in this study, a model description algorithm capable of local interpretation was applied to the inception network-derived AI to analyze what grounds they made when classifying national defense data. Specifically, we conduct a comparative analysis of explainability based on confidence values by performing LIME analysis from the Inception v2_resnet model and verify the similarity between human interpretations and LIME explanations. Furthermore, by comparing the LIME explanation results through the Top1 output results for Inception v3, Inception v2_resnet, and Xception models, we confirm the feasibility of comparing the efficiency and availability of deep learning networks using XAI.