• Journal of the Korea Convergence Society
• /
• v.10 no.5
• /
• pp.35-42
• /
• 2019

Recently, as the Internet and various wearable devices have appeared, Internet technology has contributed to obtaining more convenient information and doing business. However, as the internet is used in various parts, the attack surface points that are exposed to attacks are increasing, Attempts to invade networks aimed at taking unfair advantage, such as cyber terrorism, are also increasing. In this paper, we propose a feature selection method to improve the classification performance of the class to classify the abnormal behavior in the network traffic. The UNSW-NB15 dataset has a rare class imbalance problem with relatively few instances compared to other classes, and an undersampling method is used to eliminate it. We use the SVM, k-NN, and decision tree algorithms and extract a subset of combinations with superior detection accuracy and RMSE through training and verification. The subset has recall values of more than 98% through the wrapper based experiments and the DT_PSO showed the best performance.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.21 no.4
• /
• pp.89-100
• /
• 2011

Physical memory analysis has been an issue on a field of live forensic analysis in digital forensics until now. It is very useful to make the result of analysis more reliable, because record of user behavior and data can be founded on physical memory although process is hided. But most memory analysis focuses on windows based system. Because the diversity of target system to be analyzed rises up, it is very important to analyze physical memory based on other OS, not Windows. Mac OS X, has second market share in Operating System, is operated by loading kernel image to physical memory area. In this paper, We propose a methodology for physical memory analysis on Mac OS X using symbol information in kernel image, and acquire a process information, mounted device information, kernel information, kernel extensions(eg. KEXT) and system call entry for detecting system call hooking. In additional to the methodology, we prove that physical memory analysis is very useful though experimental study.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.17 no.1
• /
• pp.111-118
• /
• 2022

In computer vision research, the two-dimensional human pose is a very extensive research direction, especially in pose tracking and behavior recognition, which has very important research significance. The acquisition of human pose targets, which is essentially the study of how to accurately identify human targets from pictures, is of great research significance and has been a hot research topic of great interest in recent years. Human pose recognition is used in artificial intelligence on the one hand and in daily life on the other. The excellent effect of pose recognition is mainly determined by the success rate and the accuracy of the recognition process, so it reflects the importance of human pose recognition in terms of recognition rate. In this human body gesture recognition, the human body is divided into 17 key points for labeling. Not only that but also the key points are segmented to ensure the accuracy of the labeling information. In the recognition design, use the comprehensive data set MS COCO for deep learning to design a neural network model to train a large number of samples, from simple step-by-step to efficient training, so that a good accuracy rate can be obtained.

• Journal of Korean Society on Water Environment
• /
• v.38 no.2
• /
• pp.72-81
• /
• 2022

Numerous studies have investigated the occurrence and fate of microplastics in the environment; however, only limited effort has been devoted to exploring the characteristics of dissolved organic matter (DOM) leached from microplastics. In microplastic (MP)-contaminated environment, MPs are typically mixed with naturally-occurring particles, which interferes with their detection in the environment. Thus, it is necessary to distinguish between the DOM leached from MPs and those leached from natural particles and also to characterize their properties. This study investigated DOM leaching behavior from MPs (polystyrene: PS, polyvinylchloride: PVC) and natural particulates (forest soil: FS, litter leaves: LL) under light, which is considered one of the main weathering processes that affect MPs in the environment. The leached DOM concentrations and fluorescence characteristics were compared under dark versus light conditions. Regardless of the origins, UV light promoted DOM release from all the particulates. More DOM was released from natural particles than from MPs under both conditions. However, the effect of promoting DOM release by UV was more pronounced for MPs than for natural particles. It was observed from fluorescence spectra that the intensity of the humic-like region was substantially reduced when MP-derived DOM was exposed to UV light, whereas the change of intensity was very little for natural particles. Under light conditions, the ratio of protein-like to humic-like fluorescence of MP-derived DOM was higher than that of DOM from natural particles. This study implies that a substantial amount of DOM could be leached from MPs even in MP-polluted environment under UV irradiation. Protein/humic fluorescence ratio could be utilized as a fast probing indicator to separate the two sources of particles under light.

• Journal of the Korean Academy of Child and Adolescent Psychiatry
• /
• v.33 no.1
• /
• pp.16-23
• /
• 2022

Objectives: Early detection of developmental issues in infants and necessary intervention are important. To identify the comorbid conditions, a comprehensive evaluation is required. The study's objectives were to 1) generate scale items by identifying and eliciting concepts relevant to young children (12-71 months) with developmental delays, 2) develop a comprehensive screening tool for developmental delay and comorbid conditions, and 3) assess the tool's validity and cut-off. Methods: Multidisciplinary experts devised the "Infant Comprehensive Evaluation for Neurodevelopmental Delay (ICEND)," an assessment method that comes in two versions depending on the age of the child: 12-36 months and 37-71 months, through monthly seminars and focused group interviews. The ICEND is composed of three parts: risk factors, resilience factors, and clinical scales. In parts 1 and 2, there were 41 caretakers responded to the questionnaires. Part 3 involved clinicians evaluating ten subscales using 98 and 114 questionnaires for younger and older versions, respectively. The Child Behavior Checklist, Strengths and Difficulties Questionnaire, Infant-Toddler Social Emotional Assessment, and Korean Developmental Screening Test for Infants and Children were employed to analyze concurrent validity with the ICEND. The analyses were performed on both typical and high-risk infants to identify concurrent validity, reliability, and cut-off scores. Results: A total of 296 people participated in the study, with 57 of them being high-risk (19.2%). The Cronbach's alpha was positive (0.533-0.928). In the majority of domains, the ICEND demonstrated a fair discriminatory ability, with a sensitivity of 0.5-0.7 and specificity 0.7-0.9. Conclusion: The ICEND is reliable and valid, indicating its potential as an auxiliary tool for assessing neurodevelopmental delay and comorbid conditions in children aged 12-36 months and 37-71 months.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.3
• /
• pp.35-43
• /
• 2023

Emotion recognition while driving is an essential task to prevent accidents. Furthermore, in the era of autonomous driving, automobiles are the subject of mobility, requiring more emotional communication with drivers, and the emotion recognition market is gradually spreading. Accordingly, in this research plan, the driver's emotions are classified into seven categories using psychological and behavioral data, which are relatively easy to collect. The latent vectors extracted through the auto-encoder model were also used as features in this classification model, confirming that this affected performance improvement. Furthermore, it also confirmed that the performance was improved when using the framework presented in this paper compared to when the existing EEG data were included. Finally, 81% of the driver's emotion classification accuracy and 80% of F1-Score were achieved only through psychological, personal information, and behavioral data.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.2
• /
• pp.25-32
• /
• 2022

In this paper, we propose a pet monitoring system based on deep learning using an activity recognition device. The system consists of a pet's activity recognition device, a pet owner's smart device, and a server. Accelerometer and gyroscope data were collected from an Arduino-based activity recognition device, and the number of steps was calculated. The collected data is pre-processed and the amount of activity is measured by recognizing the activity in five types (sitting, standing, lying, walking, running) through a deep learning model that hybridizes CNN and LSTM. Finally, monitoring of changes in the activity, such as daily and weekly briefing charts, is provided on the pet owner's smart device. As a result of the performance evaluation, it was confirmed that specific activity recognition and activity measurement of pets were possible. Abnormal behavior detection of pets and expansion of health care services can be expected through data accumulation in the future.

• Journal of muscle and joint health
• /
• v.29 no.2
• /
• pp.100-112
• /
• 2022

Purpose: This study is a qualitative analysis of Korean-American (K-A) women's cancer prevention in the US. Methods: Qualitative research was conducted. Fifteen K-A women in four states were interviewed. Content theme analysis was used to analyze verbatim transcriptions of interviews. Results: Participants experienced difficulties in utilizing cancer screening programs. Factors include unfamiliarity with the US health care system, high health care costs or lack of health insurance, language barriers, and irregular and sporadic cancer screening participation. Participants also actively pursued non-institutional approaches to cancer prevention. They engaged in word-of-mouth informational exchanges in K-A communities, sought cancer screening in hospitals in Korea, conducted internet searches, autonomously decided on their health issues, and adopted healthy practices including better diets, physical exercise, and spiritual practices. Conclusion: It is necessary to implement measures to increase K-A women's utilization of the US cancer screening services and to encourage their active engagement in hands-on cancer prevention practices. K-A women should be empowered through increased familiarity with US cancer screening services and through the establishment of improved K-A community social services.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.21 no.3
• /
• pp.1-17
• /
• 2022

Accurate analysis of the causes of metro rail traffic congestion provides a means of addressing issues arising from metro rail traffic congestion in metropolitan areas. Currently, congestion analysis based on counting, weight detection, CCTVs, and mobile Wi-Fi is limited by poor accuracies or because studies have been restricted to single routes and trains. In this study, a train congestion analysis model was used that includes the transfer and multi-path behavior of metro passengers and train operation plans for metropolitan urban railroads. Analysis accuracy was improved by considering traffic patterns in which passengers must wait for next trains due to overcrowding. The model updates train crowding levels every 10 minutes, provides information to potential passengers, and thus, is expected to increase the social benefits provided by the Seoul metropolitan subway

• Journal of the Korean Society of Industry Convergence
• /
• v.26 no.1
• /
• pp.113-119
• /
• 2023

In this paper, we propose an optimization of a pose estimation deep learning model for recognition of abnormal behavior in unmanned stores using radio frequencies. The radio frequency use millimeter wave in the 30 GHz to 300 GHz band. Due to the short wavelength and strong straightness, it is a frequency with less grayness and less interference due to radio absorption on the object. A millimeter wave radar is used to solve the problem of personal information infringement that may occur in conventional CCTV image-based pose estimation. Deep learning-based pose estimation models generally use convolution neural networks. The convolution neural network is a combination of convolution layers and pooling layers of different types, and there are many cases of convolution filter size, number, and convolution operations, and more cases of combining components. Therefore, it is difficult to find the structure and components of the optimal posture estimation model for input data. Compared with conventional millimeter wave-based posture estimation studies, it is possible to explore the structure and components of the optimal posture estimation model for input data using genetic algorithms, and the performance of optimizing the proposed posture estimation model is excellent. Data are collected for actual unmanned stores, and point cloud data and three-dimensional keypoint information of Kinect Azure are collected using millimeter wave radar for collapse and property damage occurring in unmanned stores. As a result of the experiment, it was confirmed that the error was moored compared to the conventional posture estimation model.