• Journal of Practical Engineering Education
• /
• v.16 no.4
• /
• pp.409-414
• /
• 2024

This paper utilizes artificial intelligence to analyze vocational training evaluation data for people with disabilities and selects the optimal prediction model using various machine learning algorithms. It predicts the job categories most likely to employ trainees based on data such as gender, age, education level, type of disability, and basic learning abilities. The goal is to design customized training programs based on these predictions to enhance training efficiency and employment success rates.

• KIPS Transactions on Computer and Communication Systems
• /
• v.10 no.12
• /
• pp.329-342
• /
• 2021

The pharmaceutical industry is experiencing a productivity crisis with a low probability of success despite a long period of time and enormous cost. As a strategy to solve the productivity crisis, the use cases of Artificial Intelligence(AI) and Bigdata are increasing worldwide and tangible results are coming out. However, domestic pharmaceutical companies are taking a wait-and-see attitude to adopt AI platform for drug research. This study proposed a research model that combines the Value-based Adoption Model and the Innovation Resistance Model to empirically study the effect of value perception and resistance factors on adopting AI Platform. As a result of empirical verification, usefulness, knowledge richness, complexity, and algorithmic opacity were found to have a significant effect on perceived values. And, usefulness, knowledge richness, algorithmic opacity, trialability, technology support infrastructure were found to have a significant effect on the innovation resistance.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.32 no.5
• /
• pp.997-1008
• /
• 2022

Today, AI(Artificial Intelligence) technology is being extensively researched in various fields, including the field of malware detection. To introduce AI systems into roles that protect important decisions and resources, it must be a reliable AI model. AI model that dependent on training dataset should be verified to be robust against new attacks. Rather than generating new malware detection, attackers find malware detection that succeed in attacking by mass-producing strains of previously detected malware detection. Most of the attacks, such as adversarial attacks, that lead to misclassification of AI models, are made by slightly modifying past attacks. Robust models that can be defended against these variants is needed, and the Robustness level of the model cannot be evaluated with accuracy and recall, which are widely used as AI evaluation indicators. In this paper, we experiment a framework to evaluate robustness level by generating an adversarial sample based on one of the adversarial attacks, C&W attack, and to improve robustness level through adversarial training. Through experiments based on malware dataset in this study, the limitations and possibilities of the proposed method in the field of malware detection were confirmed.

• Smart Media Journal
• /
• v.11 no.7
• /
• pp.19-27
• /
• 2022

The artificial intelligence software market is expected to grow sixfold from 2020 to 2025. However, the software development process is not standardized and there is no standard for calculating the cost. Accordingly, each AI software development company calculates the input man-hours according to their respective development procedures and presents this as the basis for the development cost. In this study, the development stage of "artificial intelligence-based software" that learns with a large amount of data and derives and applies an algorithm was defined, and the required labor was collected by conducting a survey on the number of man-hours required for each development stage targeting developers. Correlation analysis and regression analysis were performed between the collected man-hours for each development stage, and a model for predicting the man-hours for each development stage was derived. As a result of testing the model, it showed an accuracy of 92% compared to the collected airborne effort. The man-hour prediction model proposed in this study is expected to be a tool that can be used simply for estimating man-hours and costs.

• The Journal of Korea Robotics Society
• /
• v.18 no.2
• /
• pp.216-224
• /
• 2023

Guide robots that provide services in public places have recently emerged as a non-face-to-face solution with the spread of COVID-19 and are growing. However, most guide robots provide only the same level of intelligence and the same interaction in different and changing environments. Therefore, its usefulness is limited and customers' interest is quickly lost. To solve this problem, it is necessary to develop social intelligence that can improve the robot's environment and situational awareness performance, and to continuously maintain customer interest by providing personalized and situational services. In this study, we developed guide robot services based on social HRI components that provides multi-modal context-aware. We evaluated service usefulness by measuring user satisfaction and frequency of use of the service through the survey. We analyzed the service quality attributes to identify the differentiating factors of guide robot based on social HRI components.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.95-95
• /
• 2023

Common hydrological problems of developing countries include poor data management, insufficient measuring devices and ungauged watersheds, leading to small or unreliable data availability. This has greatly affected the adoption of artificial intelligence techniques for flood risk mitigation and damage control in several developing countries. While climate datasets have recorded resounding applications, but they exhibit more uncertainties than ground-based measurements. To encourage AI adoption in developing countries with small ground-based dataset, we propose data augmentation for regression tasks and compare performance evaluation of different AI models with and without data augmentation. More focus is placed on simple models that offer lesser computational cost and higher accuracy than deeper models that train longer and consume computer resources, which may be insufficient in developing countries. To implement this approach, we modelled and predicted streamflow data of the Asa River Watershed located in Ilorin, Kwara State Nigeria. Results revealed that adequate hyperparameter tuning and proper model selection improve streamflow prediction on small water dataset. This approach can be implemented in data-scarce regions to ensure timely flood intervention and early warning systems are adopted in developing countries.

• Journal of Information Technology Services
• /
• v.22 no.4
• /
• pp.1-19
• /
• 2023

This study is a study on the effect of technical readiness factors on counselors' intention to use when applying AICC. AICC counselors experience improved customer service and emotional stability by receiving various monitor notification window services based on artificial intelligence algorithms such as customer counseling history, prohibited word control system, and customized counseling system. Accordingly, this study tried to verify using factors derived from technology readiness theory and technology acceptance theory among the factors affecting the intention to continue using AICC provided to counselors. To verify the research hypothesis, the causal relationship between variables such as Optimism, Innovativeness, Discomfort, Insecurity, and Technology Acceptance Theory, such as Team Support, Ease of Usage, and Innovation Resistance, was verified. As a result of empirical analysis, first, it was verified that Optimism has a positive (+) effect on Team Support and Ease of Usage, and Discomfort and Insecurity have a negative (-) effect on Ease of Usage and Team Support. Second, it was confirmed that Team Support and Ease of Usage had a positive effect on the Intention to use AICC. Based on the above empirical analysis results, the concepts of Technical Readiness were clearly proved, and in practical terms, AICC helped inquiry, quality evaluation, recording, and management of counseling history, ultimately increased corporate work efficiency.

• Smart Structures and Systems
• /
• v.32 no.1
• /
• pp.49-59
• /
• 2023

A bridge transportation network supplies products from various source nodes to destination nodes through bridge structures in a target region. However, recent frequent earthquakes have caused damage to bridge structures, resulting in extreme direct damage to the target area as well as indirect damage to other lifeline structures. Therefore, in this study, a surrogate model-based comprehensive framework to estimate the seismic resilience of bridge transportation networks is proposed. For this purpose, total system travel time (TSTT) is introduced for accurate performance indicator of the bridge transportation network, and an artificial neural network (ANN)-based surrogate model is constructed to reduce traffic analysis time for high-dimensional TSTT computation. The proposed framework includes procedures for constructing an ANN-based surrogate model to accelerate network performance computation, as well as conventional procedures such as direct Monte Carlo simulation (MCS) calculation and bridge restoration calculation. To demonstrate the proposed framework, Pohang bridge transportation network is reconstructed based on geographic information system (GIS) data, and an ANN model is constructed with the damage states of the transportation network and TSTT using the representative earthquake epicenter in the target area. For obtaining the seismic resilience curve of the Pohang region, five epicenters are considered, with earthquake magnitudes 6.0 to 8.0, and the direct and indirect damages of the bridge transportation network are evaluated. Thus, it is concluded that the proposed surrogate model-based framework can efficiently evaluate the seismic resilience of a high-dimensional bridge transportation network, and also it can be used for decision-making to minimize damage.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.13 no.4
• /
• pp.765-776
• /
• 2018

Artificial Intelligence is a rapidly growing technology with the latest developments in computing technology and is considered as one of the next major technologies. Chat-Bot is a system that is designed to respond to user's input according to the rules that are set up in advance and it provides more services through simple and repetitive tasks such as counseling, ordering and others. Accordingly, the study aims to present a model of a celeb-bot using Artificial Intelligence. Celeb-Bot is a combination of Celeb, which are short for Celebrity and Chat-bot. Celeb-Bot provides a Chat-Bot service that allows people to talk to a celebrity. The celeb is the best thing to build a relationship and has the advantages of being accessible to anyone. At the same time, Artificial Intelligence is a technology that can be seen as a person, not a product. Based on this, we believe that Celeb's Characteristic and Chat-bot based on artificial intelligence technologies need to be combined, so variety of products can generate synergy. It is predicted that there will be variety of derivatives that utilize this technology, and it is going to present a celeb-bot model accordingly.

• KEPCO Journal on Electric Power and Energy
• /
• v.8 no.2
• /
• pp.137-141
• /
• 2022

To inspect the power facility faults using artificial intelligence, it need that improve the accuracy of the diagnostic model are required. Data augmentation skill using generative adversarial network (GAN) is one of the best ways to improve deep learning performance. GAN model can create realistic-looking fake images using two competitive learning networks such as discriminator and generator. In this study, we intend to verify the effectiveness of virtual data generation technology by including the fake image of power facility generated through GAN in the deep learning training set. The GAN-based fake image was created for damage of LP insulator, and ResNet based normal and defect classification model was developed to verify the effect. Through this, we analyzed the model accuracy according to the ratio of normal and defective training data.