• Geomechanics and Engineering
• /
• v.25 no.1
• /
• pp.1-16
• /
• 2021

Machine learning models have been widely used for landslide susceptibility assessment (LSA) in recent years. The large number of inputs or conditioning factors for these models, however, can reduce the computation efficiency and increase the difficulty in collecting data. Feature selection is a good tool to address this problem by selecting the most important features among all factors to reduce the size of the input variables. However, two important questions need to be solved: (1) how do feature selection methods affect the performance of machine learning models? and (2) which feature selection method is the most suitable for a given machine learning model? This paper aims to address these two questions by comparing the predictive performance of 13 feature selection-based machine learning (FS-ML) models and 5 ordinary machine learning models on LSA. First, five commonly used machine learning models (i.e., logistic regression, support vector machine, artificial neural network, Gaussian process and random forest) and six typical feature selection methods in the literature are adopted to constitute the proposed models. Then, fifteen conditioning factors are chosen as input variables and 1,017 landslides are used as recorded data. Next, feature selection methods are used to obtain the importance of the conditioning factors to create feature subsets, based on which 13 FS-ML models are constructed. For each of the machine learning models, a best optimized FS-ML model is selected according to the area under curve value. Finally, five optimal FS-ML models are obtained and applied to the LSA of the studied area. The predictive abilities of the FS-ML models on LSA are verified and compared through the receive operating characteristic curve and statistical indicators such as sensitivity, specificity and accuracy. The results showed that different feature selection methods have different effects on the performance of LSA machine learning models. FS-ML models generally outperform the ordinary machine learning models. The best FS-ML model is the recursive feature elimination (RFE) optimized RF, and RFE is an optimal method for feature selection.

• Journal of Veterinary Science
• /
• v.22 no.3
• /
• pp.35.1-35.13
• /
• 2021

Background: African swine fever (ASF) is an infectious viral disease of domestic pigs that presents as a hemorrhagic fever, and for which no effective vaccine is available. The disease has a serious negative social and economic impact on pig keepers. There is limited information on the potential risk factors responsible for the spread of ASF in South Kivu. Objective: The aim of this study was to determine the potential risk factors associated with ASF infection in suspected ASF virus (ASFV)-infected pigs. Methods: We sampled whole blood from 391 pigs. Additionally, 300 pig farmers were interviewed using a structured questionnaire. Viral DNA was detected by using the real-time polymerase chain reaction technique. Results: The majority of pigs sampled, 78% (95% confidence interval [CI], 74.4-82.6), were of local breeds. Over half, 60.4% (95% CI, 55.5-65.2), were female, and most of them, 90.5% (95% CI, 87.6-93.4), were adult pigs (> 1 year old). Viral DNA was detected in 72 of the 391 sampled pigs, indicating an overall infection rate of 18.4% (95% CI, 14.5-22.4). Multivariable logistic regression analysis revealed several risk factors positively associated with ASFV infection: feeding with swill in pen (odds ratio [OR], 3.8; 95% CI, 2.12-6.77); mixed ages of pigs in the same pen (OR, 3.3; 95% CI, 1.99-5.57); introduction of new animals to the farm (OR, 5.4; 95% CI, 1.91-15.28). The risk factors that were negatively (protective) correlated with ASFV positivity were the presence of male animals and the use of an in-pen breeding system. Conclusion: Local pig farmers should be encouraged to adopt proper husbandry and feeding practices in order to increase the number of ASF-free farms.

• Journal of Urban Science
• /
• v.8 no.1
• /
• pp.51-58
• /
• 2019

This study examines the determinants of emergency care utilization and equity of access to care in elderly Koreans. Based on the data from the 2014 Korea Health Panel Survey, descriptive and logistic regression analysis was performed. The sample for this study was 1,313 individuals who participated in interviews. Predisposing factors such as age, sex, and education were significant determinants of emergency care utilization. Differences in need do not fully account for the original differences observed between subgroups of older Koreans. Health status was important determinant of older Koreans using emergency care services. Spending medical expense did not ameliorate the subgroup differences in the use of emergency care services. Nonetheless, spending medical expense remains a particularly important predictor of emergency care utilization. Health care reforms in Korea should continue to concentrate on insuring effective universal emergency care, implying that all older Koreans with need receive effective coverage. Future study is also needed to understand the access barriers that may exist for the selected demographic subgroups, i.e., those over 75, women, less educated persons, and those with higher medical expense.

• Journal of Environmental Health Sciences
• /
• v.48 no.2
• /
• pp.96-105
• /
• 2022

Background: Pyrethroid insecticides account for more than 30% of the global insecticide market and are frequently used in agricultural settings and residential and public pest control among the general population. While several animal studies have suggested that exposure to pyrethroids can alter glucose homeostasis, there is only limited evidence of the association between environmental pyrethroid exposure and diabetes in humans. Objectives: This study aimed to report environmental 3-phenoxybenzoic acid (3-PBA) concentrations in urine and evaluate its association with the risk of diabetes in Korean adults. Methods: We analyzed data from the Korean National Environmental Health Survey (KoNEHS) Cycle 2 (2012~2014) and Cycle 3 (2015~2017). A total of 10,123 participants aged ≥19 years were included. Multiple logistic regressions were used to calculate the odds ratios (ORs) for diabetes according to log-transformed urinary 3-PBA levels. We also evaluated age, sex, education, monthly income, marital status, alcohol drinking, physical activity, urinary cotinine, body mass index, and sampling season as potential effect modifiers of these associations. Results: After adjusting for all the covariates, we found significant dose-response relationships between urinary 3-PBA as quartile and the prevalence of diabetes in pooled data of KoNEHS Cycles 2 and 3. In subgroup analyses, the adverse effects of pyrethroid exposure on diabetes were significantly stronger among those aged 19~39 years (p-interaction<0.001) and those who consumed high levels of cotinine (p-interaction=0.020). Conclusions: Our findings highlight the potential diabetes risk of environmental exposure to pyrethroids and should be confirmed in large prospective studies in different populations in the future.

• Restorative Dentistry and Endodontics
• /
• v.49 no.1
• /
• pp.2.1-2.12
• /
• 2024

Objectives: This study aimed to determine the effects of 1-rooted mandibular second molar (MnSM) teeth on root canal anatomy complexities of the mandibular central incisor (MnCI), mandibular lateral incisor (MnLI), mandibular canine (MnCn), mandibular first premolar (MnFP), mandibular second premolar (MnSP), and mandibular first molar (MnFM) teeth. Materials and Methods: Cone-beam computed tomography images of 600 patients with full lower dentition were examined. Individuals with 1-rooted MnSMs were determined, and the complexity of root canal anatomy of other teeth was compared with individuals without 1-rooted MnSMs (Group-1; subjects with at least one 1-rooted MnSM, Group-2; subjects with more than a single root in both MnSMs). A second canal in MnCIs, MnLIs, MnCns, MnFPs, and MnSPs indicated a complicated root canal. The presence of a third root in MnFMs was recorded as complicated. Results: The prevalence of 1-rooted MnSMs was 12.2%, with the C-shaped root type being the most prevalent (9%). There were fewer complicated root canals in MnCIs (p = 0.02), MnLIs (p < 0.001), and MnFPs (p < 0.001) in Group 1. The other teeth showed no difference between the groups (p > 0.05). According to logistic regression analysis, 1-rooted right MnSMs had a negative effect on having complex canal systems of MnLIs and MnFPs. Left MnSMs were explanatory variables on left MnLIs and both MnFPs. Conclusions: In individuals with single-rooted MnSMs, a less complicated root canal system was observed in all teeth except the MnFMs.

• Annals of Occupational and Environmental Medicine
• /
• v.35
• /
• pp.33.1-33.9
• /
• 2023

Background: Studies on the association between occupational stress and depression have been frequently reported. However, the cross-sectional designs of studies limited insight into causal associations. In this study, we investigated the longitudinal association between occupational stress and new-onset depression among employees in a single manufacturing plant. Methods: The annual health checkup data of employees at a manufacturing plant in Korea were collected. A total of 1,837 male employees without depression who completed a health checkup during two consecutive years were included. Occupational stress was measured using a short form of the Korea Occupational Stress Scale (KOSS-SF), and depression was assessed using a Patient Health Questionnaire-2. The association between occupational stress change over the two years and newly developed depression was investigated using two logistic regression models. Results: Across all sub-factors of KOSS-SF, employees who reported increased occupational stress had a higher risk of new-onset depression. Newly developed depression was significantly associated with job demand (odds ratio [OR]: 4.34; 95% confidence interval [CI]: 2.37-7.96), job insecurity (OR: 3.21; 95% CI: 1.89-5.48), occupational climate (OR: 3.18; 95% CI: 1.91-5.31), lack of reward (OR: 2.28; 95% CI: 1.26-4.12), interpersonal conflict (OR: 2.14; 95% CI: 1.18-3.86), insufficient job control (OR: 1.93; 95% CI: 1.05-3.56), and the organizational system (OR: 1.84; 95% CI: 1.01-3.36). Conclusions: For every sub-factor of the KOSS-SF, occupational stress increase and persistent high stress were associated with the risk of developing new-onset depression. Among the seven sub-factors, job demand had the most significant effect. Our results show that occupational stress should be managed to promote employee mental healthcare.

• Annals of Occupational and Environmental Medicine
• /
• v.34
• /
• pp.17.1-17.10
• /
• 2022

Background: This study aimed to investigate the decline in quality of life (QOL) by examining changes in the employment status of workers who had completed medical treatment after an industrial accident. Methods: This study utilized the Panel Study of Worker's Compensation Insurance cohort (published in October 2020) containing a sample survey of 3,294 occupationally injured workers who completed medical care in 2017. We divided this population into four groups according to changes in working status. A multivariate logistic regression model was utilized for evaluating QOL decline by adjusting for the basic characteristics and working environment at the time of accident. Subgroup analysis evaluated whether QOL decline differed according to disability grade and industry group. Results: The QOL decline in the "maintained employment," "employed to unemployed," "remained unemployed," and "unemployed to employed" groups were 15.3%, 28.1%, 20.2%, and 11.9%, respectively. The "maintained employment" group provided a reference. As a result of adjusting for the socioeconomic status and working environment, the odds ratios (ORs) of QOL decline for the "employed to unemployed" group and the "remained unemployed" group were 2.13 (95% confidence interval [CI], 1.51-3.01) and 1.47 (95% CI, 1.13-1.90), respectively. The "unemployed to employed" group had a non-significant OR of 0.76 (95% CI, 0.54-1.07). Conclusions: This study revealed that continuous unemployment or unstable employment negatively affected industrially injured workers' QOL. Policy researchers and relevant ministries should further develop and improve "return to work" programs that could maintain decent employment avenues within the workers' compensation system.

• Journal of Distribution Science
• /
• v.11 no.6
• /
• pp.31-40
• /
• 2013

Purpose - The Northeast Asian Logistic Hub strategy was established to create a national competitive advantage in northeast Asia. Countries in this region are competing fiercely to become the central base distribution port as the volume of container shipping continues to increase due to the northeast Asian (especially Chinese) economic growth. The primary method by which shippers are improving their customer service and distribution is enhancing profits by minimizing call ports on the key route through strategic affiliations and the use of large vessels. Each nation is planning large-scale investments in the construction of sea ports that can accommodate large vessels. This paper proposes ways by which the logistical strategies of domestic corporations can keep pace with changes in government policy concerning the Northeast Asian Business Hub policy. It examines the logistics system in the Northeast Asian region, analyzes the government's Northeast Asian Business Hub policy, and suggests logistical strategies for domestic corporations through an analysis based on a questionnaire designed to grasp domestic firms' needs and goals. Research design, data and methodology - The purpose of this study is to determine how shipping companies establish partnerships with third-party logistics providers and draws out the implications of the results. The survey methods used were personal interviews and questionnaires distributed to a sample population through e-mail, fax, mail, and telephone. A total of 600 questionnaires were distributed, of which 165 were returned. Among these, ten were excluded due to insufficient content; ultimately, 155 were used for the sample. The statistical data collection process was analyzed through data coating and a statistical package program. Results - This study argues that greater flexibility in policies, administration, and systems will be needed to significantly improve established business practices. In this dissertation, we primarily identify that in order to become a center of northeast Asian logistics, Korea must adopt a new paradigm and abandon the existing systems that are based on the economic and social systems that have stemmed from bureaucracy, inflexibility, chauvinism, and equalitarianism. Flexible policies, administration, and systems will be necessary to improve business practices. Domestic corporations must establish a strategic logistics hub and related network while simultaneously pursuing value-added logistics businesses by increasing their manpower and building a logistics information system. This will strengthen their competitive edge and lead to system improvements. Conclusions - Domestic corporations must adopt a new paradigm and use more reasonable business laws, systems, and policies that are based on market-driven flexibility and transparency. Moreover, social norms and regulations should be established to help ensure political and social security. Korea must also develop a culture of tolerance for foreign companies. Finally, the paradigm defining the policy governing the development of the capital city and its satellite cities in this context must be changed.

• Journal of Gastric Cancer
• /
• v.21 no.4
• /
• pp.368-378
• /
• 2021

Purpose: When patients with early gastric cancer (EGC) undergo non-curative endoscopic submucosal dissection requiring gastrectomy (NC-ESD-RG), additional medical resources and expenses are required for surgery. To reduce this burden, predictive model for NC-ESD-RG is required. Materials and Methods: Data from 2,997 patients undergoing ESD for 3,127 forceps biopsy-proven differentiated-type EGCs (2,345 and 782 in training and validation sets, respectively) were reviewed. Using the training set, the logistic stepwise regression analysis determined the independent predictors of NC-ESD-RG (NC-ESD other than cases with lateral resection margin involvement or piecemeal resection as the only non-curative factor). Using these predictors, a risk-scoring system for predicting NC-ESD-RG was developed. Performance of the predictive model was examined internally with the validation set. Results: Rate of NC-ESD-RG was 17.3%. Independent pre-ESD predictors for NC-ESD-RG included moderately differentiated or papillary EGC, large tumor size, proximal tumor location, lesion at greater curvature, elevated or depressed morphology, and presence of ulcers. A risk-score was assigned to each predictor of NC-ESD-RG. The area under the receiver operating characteristic curve for predicting NC-ESD-RG was 0.672 in both training and validation sets. A risk-score of 5 points was the optimal cut-off value for predicting NC-ESD-RG, and the overall accuracy was 72.7%. As the total risk score increased, the predicted risk for NC-ESD-RG increased from 3.8% to 72.6%. Conclusions: We developed and validated a risk-scoring system for predicting NC-ESD-RG based on pre-ESD variables. Our risk-scoring system can facilitate informed consent and decision-making for preoperative treatment selection between ESD and surgery in patients with EGC.

• Journal of Intelligence and Information Systems
• /
• v.18 no.1
• /
• pp.125-141
• /
• 2012

These days, the malicious attacks and hacks on the networked systems are dramatically increasing, and the patterns of them are changing rapidly. Consequently, it becomes more important to appropriately handle these malicious attacks and hacks, and there exist sufficient interests and demand in effective network security systems just like intrusion detection systems. Intrusion detection systems are the network security systems for detecting, identifying and responding to unauthorized or abnormal activities appropriately. Conventional intrusion detection systems have generally been designed using the experts' implicit knowledge on the network intrusions or the hackers' abnormal behaviors. However, they cannot handle new or unknown patterns of the network attacks, although they perform very well under the normal situation. As a result, recent studies on intrusion detection systems use artificial intelligence techniques, which can proactively respond to the unknown threats. For a long time, researchers have adopted and tested various kinds of artificial intelligence techniques such as artificial neural networks, decision trees, and support vector machines to detect intrusions on the network. However, most of them have just applied these techniques singularly, even though combining the techniques may lead to better detection. With this reason, we propose a new integrated model for intrusion detection. Our model is designed to combine prediction results of four different binary classification models-logistic regression (LOGIT), decision trees (DT), artificial neural networks (ANN), and support vector machines (SVM), which may be complementary to each other. As a tool for finding optimal combining weights, genetic algorithms (GA) are used. Our proposed model is designed to be built in two steps. At the first step, the optimal integration model whose prediction error (i.e. erroneous classification rate) is the least is generated. After that, in the second step, it explores the optimal classification threshold for determining intrusions, which minimizes the total misclassification cost. To calculate the total misclassification cost of intrusion detection system, we need to understand its asymmetric error cost scheme. Generally, there are two common forms of errors in intrusion detection. The first error type is the False-Positive Error (FPE). In the case of FPE, the wrong judgment on it may result in the unnecessary fixation. The second error type is the False-Negative Error (FNE) that mainly misjudges the malware of the program as normal. Compared to FPE, FNE is more fatal. Thus, total misclassification cost is more affected by FNE rather than FPE. To validate the practical applicability of our model, we applied it to the real-world dataset for network intrusion detection. The experimental dataset was collected from the IDS sensor of an official institution in Korea from January to June 2010. We collected 15,000 log data in total, and selected 10,000 samples from them by using random sampling method. Also, we compared the results from our model with the results from single techniques to confirm the superiority of the proposed model. LOGIT and DT was experimented using PASW Statistics v18.0, and ANN was experimented using Neuroshell R4.0. For SVM, LIBSVM v2.90-a freeware for training SVM classifier-was used. Empirical results showed that our proposed model based on GA outperformed all the other comparative models in detecting network intrusions from the accuracy perspective. They also showed that the proposed model outperformed all the other comparative models in the total misclassification cost perspective. Consequently, it is expected that our study may contribute to build cost-effective intelligent intrusion detection systems.