• Journal of Korea Society of Digital Industry and Information Management
• /
• v.19 no.4
• /
• pp.13-27
• /
• 2023

In this study, we tried to identify research trends on artificial intelligence security focusing on the United States, United Kingdom, and South Korea. In Elsevier's Scopus We collected 4,983 papers related to artificial intelligence security published from 2018 to 2022 and by using the abstracts of the collected papers, Keyword frequency and centrality analysis were conducted. By calculating keyword frequency, keywords with high frequency of appearance were identified and through the centrality analysis, central research keywords were identified by country. Through the analysis results, research related to artificial intelligence, machine learning, Internet of Things, and cybersecurity in each country was conducted as the most central and highly mediating research. The implication for Korea is that research related to cybersecurity, privacy, and anomaly detection has lower centralities compared to the United States and research related to big data has lower centralities compared to United Kingdom. Therefore, various researches that intensively apply artificial intelligence technology to these fields are needed.

• Korean Journal of Artificial Intelligence
• /
• v.5 no.1
• /
• pp.10-17
• /
• 2017

This study is the first attempt to raise face recognition ability through machine learning algorithm and apply to CRM's information gathering, analysis and application. In other words, through face recognition of VIP customer in distribution field, we can proceed more prompt and subdivided customized services. The interest in machine learning, which is used to implement artificial intelligence, has increased, and it has become an age to automate it by using machine learning beyond the way that a person directly models an object recognition process. Among them, Deep Learning is evaluated as an advanced technology that shows amazing performance in various fields, and is applied to various fields of image recognition. Face recognition, which is widely used in real life, has been developed to recognize criminals' faces and catch criminals. In this study, two image analysis models, TF-SLIM and Inception-V3, which are likely to be used for criminal face recognition, were selected, analyzed, and implemented. As an evaluation criterion, the image recognition model was evaluated based on the accuracy of the face recognition program which is already being commercialized. In this experiment, it was evaluated that the recognition accuracy was good when the accuracy of the image classification was more than 90%. A limit of our study which is a way to raise face recognition is left as a further research subjects.

• KIPS Transactions on Software and Data Engineering
• /
• v.9 no.9
• /
• pp.267-280
• /
• 2020

Mirror is a physical reflective surface, typically of glass coated with a metal amalgam, and it is to reflect an image clearly. They are available everywhere anytime and become an essential tool for us to observe our faces and appearances. With the advent of modern software technology, we are motivated to enhance the reflection capability of mirrors with the convenience and intelligence of realtime processing, microservices, and machine learning. In this paper, we present a development of Digital Mirror System that provides the realtime reflection functionality as mirror while providing additional convenience and intelligence including personal information retrieval, public information retrieval, appearance age detection, and emotion detection. Moreover, it provides a multi-model user interface of touch-based, voice-based, and gesture-based. We present our design and discuss how it can be implemented with current technology to deliver the realtime mirror reflection while providing useful information and machine learning intelligence.

• Convergence Security Journal
• /
• v.22 no.3
• /
• pp.101-110
• /
• 2022

As the cases of using artificial intelligence in various fields increase, attempts to solve various issues through artificial intelligence in the intrusion detection field are also increasing. However, the black box basis, which cannot explain or trace the reasons for the predicted results through machine learning, presents difficulties for security professionals who must use it. To solve this problem, research on explainable AI(XAI), which helps interpret and understand decisions in machine learning, is increasing in various fields. Therefore, in this paper, we propose an explanatory AI to enhance the reliability of machine learning-based intrusion detection prediction results. First, the intrusion detection model is implemented through XGBoost, and the description of the model is implemented using SHAP. And it provides reliability for security experts to make decisions by comparing and analyzing the existing feature importance and the results using SHAP. For this experiment, PKDD2007 dataset was used, and the association between existing feature importance and SHAP Value was analyzed, and it was verified that SHAP-based explainable AI was valid to give security experts the reliability of the prediction results of intrusion detection models.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.47 no.1
• /
• pp.41-50
• /
• 2024

Korea is facing a significant problem with historically low fertility rates, which is becoming a major social issue affecting the economy, labor force, and national security. This study analyzes the factors contributing to the regional gap in fertility rates and derives policy implications. The government and local authorities are implementing a range of policies to address the issue of low fertility. To establish an effective strategy, it is essential to identify the primary factors that contribute to regional disparities. This study identifies these factors and explores policy implications through machine learning and explainable artificial intelligence. The study also examines the influence of media and public opinion on childbirth in Korea by incorporating news and online community sentiment, as well as sentiment fear indices, as independent variables. To establish the relationship between regional fertility rates and factors, the study employs four machine learning models: multiple linear regression, XGBoost, Random Forest, and Support Vector Regression. Support Vector Regression, XGBoost, and Random Forest significantly outperform linear regression, highlighting the importance of machine learning models in explaining non-linear relationships with numerous variables. A factor analysis using SHAP is then conducted. The unemployment rate, Regional Gross Domestic Product per Capita, Women's Participation in Economic Activities, Number of Crimes Committed, Average Age of First Marriage, and Private Education Expenses significantly impact regional fertility rates. However, the degree of impact of the factors affecting fertility may vary by region, suggesting the need for policies tailored to the characteristics of each region, not just an overall ranking of factors.

• Soft Computing and Machine Intelligence
• /
• v.1 no.1
• /
• pp.24-30
• /
• 2021

The accelerated growth of the internet and the enormous amount of data availability has become the primary reason for machine learning applications for data analysis and, more specifically, pattern recognition and decision making. In this paper, we focused on the crowdfunding site Kickstarter and collected the comments in order to apply neural networks to classify the projects based on the sentiments of backers. The power of customer reviews and sentiment analysis has motivated us to apply this technique in crowdfunding to find timely indications and identify suspicious activities and mitigate the risk of money loss.

• International Journal of Internet, Broadcasting and Communication
• /
• v.13 no.2
• /
• pp.7-13
• /
• 2021

Machine vision is a technology that helps the computer as if a person recognizes and determines things. In recent years, as advanced technologies such as optical systems, artificial intelligence and big data advanced in conventional machine vision system became more accurate quality inspection and it increases the manufacturing efficiency. In machine vision systems using deep learning, the image quality of the input image is very important. However, most images obtained in the industrial field for quality inspection typically contain noise. This noise is a major factor in the performance of the machine vision system. Therefore, in order to improve the performance of the machine vision system, it is necessary to eliminate the noise of the image. There are lots of research being done to remove noise from the image. In this paper, we propose an autoencoder based machine vision system to eliminate noise in the image. Through experiment proposed model showed better performance compared to the basic autoencoder model in denoising and image reconstruction capability for MNIST and fashion MNIST data sets.

• Journal of Radiation Protection and Research
• /
• v.46 no.4
• /
• pp.204-212
• /
• 2021

Background: Identification of radioisotopes for plastic scintillation detectors is challenging because their spectra have poor energy resolutions and lack photo peaks. To overcome this weakness, many researchers have conducted radioisotope identification studies using machine learning algorithms; however, the effect of data normalization on radioisotope identification has not been addressed yet. Furthermore, studies on machine learning-based radioisotope identifiers for plastic scintillation detectors are limited. Materials and Methods: In this study, machine learning-based radioisotope identifiers were implemented, and their performances according to data normalization methods were compared. Eight classes of radioisotopes consisting of combinations of ²²Na, ⁶⁰Co, and ¹³⁷Cs, and the background, were defined. The training set was generated by the random sampling technique based on probabilistic density functions acquired by experiments and simulations, and test set was acquired by experiments. Support vector machine (SVM), artificial neural network (ANN), and convolutional neural network (CNN) were implemented as radioisotope identifiers with six data normalization methods, and trained using the generated training set. Results and Discussion: The implemented identifiers were evaluated by test sets acquired by experiments with and without gain shifts to confirm the robustness of the identifiers against the gain shift effect. Among the three machine learning-based radioisotope identifiers, prediction accuracy followed the order SVM > ANN > CNN, while the training time followed the order SVM > ANN > CNN. Conclusion: The prediction accuracy for the combined test sets was highest with the SVM. The CNN exhibited a minimum variation in prediction accuracy for each class, even though it had the lowest prediction accuracy for the combined test sets among three identifiers. The SVM exhibited the highest prediction accuracy for the combined test sets, and its training time was the shortest among three identifiers.

• International Journal of Computer Science & Network Security
• /
• v.22 no.2
• /
• pp.318-326
• /
• 2022

Marketing stress testing is a new method of identifying the company's strengths and weaknesses in a turbulent environment. Technically, this is a complex procedure, so it involves artificial intelligence and machine learning. The main problem is currently the development of methodological approaches to the development of the company's digital model, which will provide a framework for machine learning. The aim of the study was to identify and develop an author's approach to the parameterization of the company's business processes for machine learning-based marketing stress testing. This aim provided the company's activities to be considered as a set of elements (business processes, products) and factors that affect them (marketing environment). The article proposes an author's approach to the parameterization of the company's business processes for machine learning-based marketing stress testing. The proposed approach includes four main elements that are subject to parameterization: elements of the company's internal environment, factors of the marketing environment, the company' core competency and factors impacting the company. Matrices for evaluating the results of the work of expert groups to determine the degree of influence of the marketing environment factors were developed. It is proposed to distinguish between mega-level, macro-level, meso-level and micro-level factors depending on the degree of impact on the company. The methodological limitation of the study is that it involves the modelling method as the only one possible at this stage of the study. The implementation limitation is that the proposed approach can only be used if the company plans to use machine learning for marketing stress testing.

• Korean Circulation Journal
• /
• v.53 no.10
• /
• pp.677-689
• /
• 2023

Background and Objectives: There is limited evidence regarding machine-learning prediction for the recurrence of atrial fibrillation (AF) after electrical cardioversion (ECV). This study aimed to predict the recurrence of AF after ECV using machine learning of clinical features and electrocardiograms (ECGs) in persistent AF patients. Methods: We analyzed patients who underwent successful ECV for persistent AF. Machine learning was designed to predict patients with 1-month recurrence. Individual 12-lead ECGs were collected before and after ECV. Various clinical features were collected and trained the extreme gradient boost (XGBoost)-based model. Ten-fold cross-validation was used to evaluate the performance of the model. The performance was compared to the C-statistics of the selected clinical features. Results: Among 718 patients (mean age 63.5±9.3 years, men 78.8%), AF recurred in 435 (60.6%) patients after 1 month. With the XGBoost-based model, the areas under the receiver operating characteristic curves (AUROCs) were 0.57, 0.60, and 0.63 if the model was trained by clinical features, ECGs, and both (the final model), respectively. For the final model, the sensitivity, specificity, and F1-score were 84.7%, 28.2%, and 0.73, respectively. Although the AF duration showed the best predictive performance (AUROC, 0.58) among the clinical features, it was significantly lower than that of the final machine-learning model (p<0.001). Additional training of extended monitoring data of 15-minute single-lead ECG and photoplethysmography in available patients (n=261) did not significantly improve the model's performance. Conclusions: Machine learning showed modest performance in predicting AF recurrence after ECV in persistent AF patients, warranting further validation studies.