• Journal of Korean society of Dental Hygiene
• /
• v.22 no.4
• /
• pp.289-295
• /
• 2022

Objectives: In this study, raw data from the 7th (2016-2018) of the Korea National Health and Nutrition Examination Survey were used, and a total of 2,430 people were selected as participants to analyze the factors related to unmet dental care needs in postmenopausal women. Methods: Frequency analysis, cross analysis, and logistic regression analysis were performed for general characteristics, oral health characteristics, and unmet dental care needs related factors using IBM SPSS Statistics 21.0 program. Results: The unmet dental care related factors were 1.527 times higher in the case of not having oral examinations than in the case of having oral examinations. For those who did not make use of the dental clinic, 8.667 times, 2.913 times for bad oral health, and 1.912 times for usually showed that unmet dental care was higher. Inconvenience with speaking was 1.578 times higher, and in the absence of implants, unmet dental care needs was 1.510 times higher. In the case of no chewing difficulty, was 0.380 times lower. Conclusions: Based on the above results, in order to achieve the policy goal to reduce unmet dental care needs, policy support and interest are needed above all to accurately identify and solve specific problems.

• Asia-Pacific Journal of Business
• /
• v.12 no.4
• /
• pp.13-40
• /
• 2021

Purpose - The purpose of this study is to predict corporate corruption in emerging markets such as Brazil, Russia, India, and China (BRIC) using different machine learning techniques. Since corruption is a significant problem that can affect corporate performance, particularly in emerging markets, it is important to correctly identify whether a company engages in corrupt practices. Design/methodology/approach - In order to address the research question, we employ predictive analytic techniques (machine learning methods). Using the World Bank Enterprise Survey Data, this study evaluates various predictive models generated by seven supervised learning algorithms: k-Nearest Neighbour (k-NN), Naïve Bayes (NB), Decision Tree (DT), Decision Rules (DR), Logistic Regression (LR), Support Vector Machines (SVM), and Artificial Neural Network (ANN). Findings - We find that DT, DR, SVM and ANN create highly accurate models (over 90% of accuracy). Among various factors, firm age is the most significant, while several other determinants such as source of working capital, top manager experience, and the number of permanent full-time employees also contribute to company corruption. Research implications or Originality - This research successfully demonstrates how machine learning can be applied to predict corporate corruption and also identifies the major causes of corporate corruption.

• Asia-Pacific Journal of Business
• /
• v.13 no.2
• /
• pp.107-122
• /
• 2022

Purpose - This study aims to develop machine learning models to predict administrative issue designation in KOSDAQ Market using financial data. Design/methodology/approach - Employing four classification techniques including logistic regression, support vector machine, random forest, and gradient boosting to a matched sample of five hundred and thirty-six firms over an eight-year period, the authors develop prediction models and explore the practicality of the models. Findings - The resulting four binary selection models reveal overall satisfactory classification performance in terms of various measures including AUC (area under the receiver operating characteristic curve), accuracy, F1-score, and top quartile lift, while the ensemble models (random forest and gradienct boosting) outperform the others in terms of most measures. Research implications or Originality - Although the assessment of administrative issue potential of firms is critical information to investors and financial institutions, detailed empirical investigation has lagged behind. The current research fills this gap in the literature by proposing parsimonious prediction models based on a few financial variables and validating the applicability of the models.

• International Journal of Computer Science & Network Security
• /
• v.23 no.2
• /
• pp.13-30
• /
• 2023

Since the COVID-19 vaccine became available, people have been sharing their opinions on social media about getting vaccinated, causing discussions of the vaccine to trend on Twitter alongside certain events, making the website a rich data source. This paper explores people's perceptions regarding the COVID-19 vaccine during certain events and how these events influenced public opinion about the vaccine. The data consisted of tweets sent during seven important events that were gathered within 14 days of the first announcement of each event. These data represent people's reactions to these events without including irrelevant tweets. The study targeted tweets sent in Arabic from users located in Saudi Arabia. The data were classified as positive, negative, or neutral in tone. Four classifiers were used-support vector machine (SVM), naïve Bayes (NB), logistic regression (LOGR), and random forest (RF)-in addition to a deep learning model using BiLSTM. The results showed that the SVM achieved the highest accuracy, at 91%. Overall perceptions about the COVID-19 vaccine were 54% negative, 36% neutral, and 10% positive.

• Structural Monitoring and Maintenance
• /
• v.9 no.4
• /
• pp.337-357
• /
• 2022

The objective of this study is to present a data-driven machine learning (ML) framework for predicting ultimate shear strength and failure modes of reinforced concrete ledge beams. Experimental tests were collected on these beams with different loading, geometric and material properties. The database was analyzed using different ML algorithms including decision trees, discriminant analysis, support vector machine, logistic regression, nearest neighbors, naïve bayes, ensemble and artificial neural networks to identify the governing and critical parameters of reinforced concrete ledge beams. The results showed that ML framework can effectively identify the failure mode of these beams either web shear failure, flexural failure or ledge failure. ML framework can also derive equations for predicting the ultimate shear strength for each failure mode. A comparison of the ultimate shear strength of ledge failure was conducted between the experimental results and the results from the proposed equations and the design equations used by international codes. These comparisons indicated that the proposed ML equations predict the ultimate shear strength of reinforced concrete ledge beams better than the design equations of AASHTO LRFD-2020 or PCI-2020.

• International conference on construction engineering and project management
• /
• 2009.05a
• /
• pp.1260-1265
• /
• 2009

Topographical features in South Korea is characterized that 70% of territory is composed of the mountains that can experience intense rainfall during storms in the summer and autumn. Efficient planning and management of landscape becomes utmost important since the cutting slopes in the mountain areas have been increased due to the limited construction areas for the roadway and residential development. This paper proposed an efficient way of slope management for the landslide risk by developing Web-GIS landslide risk management system. By deploying the Logistic Regression Analysis, the system could increase the prediction accuracy that the landslide disaster might be occurred. High resolution survey technology using GPS and Total-Station could extract the exact position and visual shape of the slopes that accurately describe the slope information. Through the proposed system, the prediction of damage areas from the landslide could also make it easy to efficiently identify the level of landslide risks via web-based user interface. It is expected that the proposed landslide risk management system can support the decision making framework during the identification, prediction, and management of the landslide risks.

• Korean Journal of Health Education and Promotion
• /
• v.3 no.1
• /
• pp.83-98
• /
• 2001

Although periodic health examination has been one of the most common practices of preventive medicine, its effect on modification of risk behavior has been seldom assessed. Thus, this study attempted to demonstrate the influence of a health examination on modification of cardiovascular disease related health risk behaviors such as smoking, physical inactivity, and obesity. Data of 893 adults were derived from two types of a popular and highly acclaimed health examination program. With a conceptual model constructed using Persuasive Communication variables, McNemar tests examined Source-Outcome association, hypothesizing that different health examination programs would yield different levels of behavior change in smoking, physical inactivity, and obesity. No significant behavior change was found in any of the two health examination programs. Instead, previously established Receiver-Outcome relationship was reconfirmed by logistic regression modeling where gender was the most prominent predictor of all three behaviors. Men were more likely to be current smokers (OR=0.029), exercisers (OR=2.629), and obese (OR=0.237). The importance of followups after health examination is highly stressed as well as that of gender-specific health education strategies. This study recommends applying the social-ecological approaches in health examination, which emphasizes the support and collaboration at individual, family, organizations, community, and policy level to improve health. Long term and qualitative evaluation of health examination may provide more foundation for increasing the effectiveness of health education and communication in health examinations.

• ETRI Journal
• /
• v.45 no.1
• /
• pp.105-118
• /
• 2023

We tested the feasibility of automated discrimination of patients with panic disorder (PD) from healthy controls (HCs) based on multimodal physiological responses using machine learning. Electrocardiogram (ECG), electrodermal activity (EDA), respiration (RESP), and peripheral temperature (PT) of the participants were measured during three experimental phases: rest, stress, and recovery. Eleven physiological features were extracted from each phase and used as input data. Logistic regression (LoR), k-nearest neighbor (KNN), support vector machine (SVM), random forest (RF), and multilayer perceptron (MLP) algorithms were implemented with nested cross-validation. Linear regression analysis showed that ECG and PT features obtained in the stress and recovery phases were significant predictors of PD. We achieved the highest accuracy (75.61%) with MLP using all 33 features. With the exception of MLP, applying the significant predictors led to a higher accuracy than using 24 ECG features. These results suggest that combining multimodal physiological signals measured during various states of autonomic arousal has the potential to differentiate patients with PD from HCs.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.467-468
• /
• 2021

In order to improve the performance of the indoor positioning system using BLE beacons, a receiver that measures the angle of arrival among the direction finding technologies supported by BLE5.1 was manufactured and analyzed by machine learning to measure the optimal position. For the creation and testing of machine learning models, k-nearest neighbor classification and regression, logistic regression, support vector machines, decision tree artificial neural networks, and deep neural networks were used to learn and test. As a result, when the test set 4 produced in the study was used, the accuracy was up to 99%.

• The Journal of Information Systems
• /
• v.32 no.2
• /
• pp.87-108
• /
• 2023

Purpose The main purpose of this study is to improve fake news detection performance by using video information to overcome the limitations of extant text- and image-oriented studies that do not reflect the latest news consumption trend. Design/methodology/approach This study collected video clips and related information including news scripts, speakers' facial expression, and video metadata from YouTube to develop fake news detection model. Based on the collected data, seven combinations of related information (i.e. scripts, video metadata, facial expression, scripts and video metadata, scripts and facial expression, and scripts, video metadata, and facial expression) were used as an input for taining and evaluation. The input data was analyzed using six models such as support vector machine and deep neural network. The area under the curve(AUC) was used to evaluate the performance of classification model. Findings The results showed that the ACU and accuracy values of three features combination (scripts, video metadata, and facial expression) were the highest in logistic regression, naïve bayes, and deep neural network models. This result implied that the fake news detection could be improved by using video information(video metadata and facial expression). Sample size of this study was relatively small. The generalizablity of the results would be enhanced with a larger sample size.