• The Korean Journal of Applied Statistics
• /
• v.29 no.2
• /
• pp.309-319
• /
• 2016

For binary classification models, we consider a risk score that is a function of linear scores and estimate the coefficients of the linear scores. There are two estimation methods: one is to obtain MLEs using logistic models and the other is to estimate by maximizing AUC. AUC approach estimates are better than MLEs when using logistic models under a general situation which does not support logistic assumptions. This paper considers imbalanced data that contains a smaller number of observations in the default class than those in the non-default for credit assessment models; consequently, the AUC approach is applied to imbalanced data. Various logit link functions are used as a link function to generate imbalanced data. It is found that predicted coefficients obtained by the AUC approach are equivalent to (or better) than those from logistic models for low default probability - imbalanced data.

• Journal of the Korean Society of Radiology
• /
• v.13 no.4
• /
• pp.539-546
• /
• 2019

Liver biopsy is invasive and it is a risk of complications. Nevertheless, liver biopsy is gold standard for predicting liver fibrosis. To compensate for these shortcomings, in this study, the liver fibrosis stage was divided using Fibroscan(R) in 200 chronic hepatitis C patients. And, the usefulness and cut-off values of fibrosis index based on four factors(FIB-4), AST to platelet ratio index(APRI) and AST/ALT ratio(AAR) calculated as serum tests were investigated by analyzing ROC curve. As a result, using FIB-4 and APRI rather than AAR is appropriate for evaluation of liver fibrosis. And using APRI to predict significant Fibrosis(F2) and FIB-4 is considered useful for predicting cirrhosis(F4). By applying the advantages of the serum based liver fibrosis marker, which are convenient and repeatable, liver fibrosis follow-up term can be reduced, and furthermore, the prevalence of liver cirrhosis and hepatocellular carcinoma(HCC) can be reduced.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.19 no.1
• /
• pp.29-35
• /
• 2019

As the fourth industrial revolution is emerging, many companies are increasingly interested in smart factories and the importance of sensors is being emphasized. In the case that sensors for collecting sensing data fail, the plant could not be optimized and further it could not be operated properly, which may incur a financial loss. For this purpose, it is necessary to diagnose the status of sensors to prevent sensor' fault. In the paper, we propose a scheme to diagnose digital-sensor' fault by analyzing the rising time and falling time of digital sensors through the LSTM(Long Short Term Memory) of Deep Learning RNN algorithm. Experimental results of the proposed scheme are compared with those of rule-based fault diagnosis algorithm in terms of AUC(Area Under the Curve) of accuracy and ROC(Receiver Operating Characteristic) curve. Experimental results show that the proposed system has better and more stable performance than the rule-based fault diagnosis algorithm.

• Journal of the Korea Convergence Society
• /
• v.11 no.12
• /
• pp.283-290
• /
• 2020

The purpose of this study was to evaluate the usefulness of the triglyceride and glucose(TyG) index to predict the risk of hyperuricemia in Korean adults. This study included 14,266 men and 9,033 women over 20 years old who underwent health screenings from 2017 to 2019 at a general hospital in Seoul. To confirm the risk of hyperuricemia and predictive ability of the TyG index, logistic regression analysis and ROC curves were obtained. The accuracy of the TyG index for predicting hyperuricemia was 0.68, 0.61 for men and 0.67 for women(respectively p<0.001). The risk of hyperuricemia in the TyG index was 1.69 times higher in the fourth quartile than in the first quartile, 2.03 times higher in men and 2.07 times higher in women(respectively p<0.05). Thus the TyG index was not of high diagnostic usefulness as a screening test for hyperuricemia, but it was related to the TyG index and hyperuricemia.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.7
• /
• pp.9-17
• /
• 2021

In this study, we propose a part of the CDSS(Clinical Decision Support System) study, a system that can classify chemotherapy, one of the treatment methods for colorectal cancer patients. In the treatment of colorectal cancer, the selection of chemotherapy according to the patient's condition is very important because it is directly related to the patient's survival period. Therefore, in this study, chemotherapy was classified using a machine learning algorithm by creating a baseline model, a pathological model, and a combined model using both characteristics of the patient using the individual and pathological characteristics of colorectal cancer patients. As a result of comparing the prediction accuracy with Top-n Accuracy, ROC curve, and AUC, it was found that the combined model showed the best prediction accuracy, and that the LGBM algorithm had the best performance. In this study, a chemotherapy classification model suitable for the patient's condition was constructed by classifying the model by patient characteristics using a machine learning algorithm. Based on the results of this study in future studies, it will be helpful for CDSS research by creating a better performing chemotherapy classification model.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.33 no.4
• /
• pp.687-697
• /
• 2023

Botnets, whose attack patterns are becoming more sophisticated and diverse, are recognized as one of the most serious cybersecurity threats today. This paper revisits the experimental results of botnet detection using autoencoder, a semi-supervised deep learning model, for UGR and CTU-13 data sets. To prepare the input vectors of autoencoder, we create data points by grouping the NetFlow records into sliding windows based on source IP address and aggregating them to form features. In particular, we discover a simple power-law; that is the number of data points that have some flow-degree is proportional to the number of NetFlow records aggregated in them. Moreover, we show that our power-law fits the real data very well resulting in correlation coefficients of 97% or higher. We also show that this power-law has an impact on the learning of autoencoder and, as a result, influences the performance of botnet detection. Furthermore, we evaluate the performance of autoencoder using the area under the Receiver Operating Characteristic (ROC) curve.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.3
• /
• pp.107-114
• /
• 2022

In this paper, we have enhanced the risk prediction of hypertension using the feature selection method in the Korean National Health and Nutrition Examination Survey (KNHANES) database of the Korea Centers for Disease Control and Prevention. The study identified various risk factors correlated with chronic hypertension. The paper is divided into three parts. Initially, the data preprocessing step of removes missing values, and performed z-transformation. The following is the feature selection (FS) step that used a factor analysis (FA) based on the feature selection method in the dataset, and feature importance (FI) and multicollinearity analysis (MC) were compared based on FS. Finally, in the predictive analysis stage, it was applied to detect and predict the risk of hypertension. In this study, we compare the accuracy, f-score, area under the ROC curve (AUC), and mean standard error (MSE) for each model of classification. As a result of the test, the proposed MC-FA-RF model achieved the highest accuracy of 80.12%, MSE of 0.106, f-score of 83.49%, and AUC of 85.96%, respectively. These results demonstrate that the proposed MC-FA-RF method for hypertension risk predictions is outperformed other methods.

• Journal of Korean Clinical Nursing Research
• /
• v.30 no.1
• /
• pp.35-44
• /
• 2024

Purpose: The purpose of this study is to compare the predictive validity of pressure injury risk assessment, Braden, Braden Q and Braden QD for pediatric patients. Methods: Prospective observational study included patients under the age of 19 who were hospitalized to general wards, intensive care units of a children's hospital. Characteristics related to pressure injury were collected, and predicted validity was compared by calculating the areas under the curve (AUC) of the Braden, Braden Q, and Braden QD scales. Results: A total of 689 patients were included in the study. A total of 13 (1.9%) patients had pressure injuries, and the number of pressure injuries was 17. Factors related to the occurrence of pressure injuries were 9 (52.9%) immobility-related and 8 (47.1%) medical device-related. The AUC for each scale was .91 (95% CI .89~.94) for Braden, .92 (95% CI .90~.95) for Braden Q, and .94(95% CI .92~.96) for Braden QD. The optimal cut-off points were identified as 16 for Braden (sensitivity=88.8%, specificity=86.4%), 17 for Braden Q(sensitivity=63.6%, specificity=94.9%), and 12 for Braden QD (sensitivity=94.4%, specificity=88.7%). Conclusion: The Braden QD scale demonstrated the highest predictive validity for pressure injuries in pediatric patients and is expected to be valuable tool in preventing pediatrics pressure injuries.

• Journal of Preventive Medicine and Public Health
• /
• v.32 no.3
• /
• pp.361-373
• /
• 1999

Objectives: Receive Operating Characteristic(ROC) curve with the area under the ROC curve(AUC) is one of the most popular indicator to evaluate the criterion validity of the measurement tool. This study was conducted to develop a standardized questionnaire to discriminate workers at high-risk of work-related musculoskeletal disorders using ROC analysis. Methods: The diagnostic results determined by rehabilitation medicine specialists in 370 persons(89 shipyard CAD workers, 113 telephone directory assistant operators, 79 women with occupation, and 89 housewives) were compared with participant's own replies to 'the questionnair on the worker's subjective physical symptoms'(Kwon, 1996). The AUC's from four models with different methods in item selection and weighting were compared with each other. These 4 models were applied to 225 persons, working in an assembly line of motor vehicle, for the purpose of AUC reliability test. Results: In a weighted model with 11 items, the AUC was 0.8155 in the primary study population, and 0.8026 in the secondary study population(p=0.3780). It was superior in the aspects of discriminability, reliability and convenience. A new questionnaire of musculoskeletal disorder could be constructed by this model. Conclusion: A more valid questionnaire with a small number of items and the quantitative weight scores useful for the relative comparisons are the main results of this study. While the absolute reference value applicable to the wide range of populations was not estimated, the basic intent of this study, developing a surveillance fool through quantitative validation of the measures, would serve for the systematic disease prevention activities.

• The Korean Journal of Nuclear Medicine Technology
• /
• v.13 no.3
• /
• pp.24-30
• /
• 2009

Purpose: Today, Prostate cancer has been gradually increasing, according to the change of internal incidence rate of cancer. Generally, prostate cancer has lead to dead over 90%, in case of metastasis of lymph node and bone. So, innovative development of new radiopharmaceutical and imaging modality is progressed for detection of that metastasis, in nuclear medicine, now. Therefore, this study shows the usefulness of $^{18}F$-Fluoride PET/CT improved diagnosability on bone metastasis of prostate cancer. Materials and Methods: In this study, 33 male patients with prostate cancer were examined (The mean age: $67.8{\pm}10.2$ years old). Every patient was done each whole body bone scan (WBBS) and $^{18}F$-Fluoride positron emission tomography/computed tomography ($^{18}F$-Fluoride PET/CT). And then, using Receiver Operating Characteristic Curve (ROC curve), each sensitivity and specificity of two modalities was measured and compared with. Results: In 22 patients (66.6%) of all, bone metastasis was detected. And, in WBBS, sensitivity was 63.6%, specificity, 81.8%; in $^{18}F$-Fluoride PET/CT, sensitivity was 100% and specificity was 90.9%. As a result of ROC curve, AUROC (The Area under an ROC) of WBBS was 0.778, and that of $^{18}F$-Fluoride PET/CT, 0.942. Conclusions: $^{18}F$-Fluoride PET/CT was higher both sensitivity and specificity than WBBS, and it was valuable to detect bone metastasis of prostate cancer more definitely, with 3D imaging realization. Also, in $^{18}F$-Fluoride PET/CT, physiological images were acquired in more short time than WBBS, so, it was possible to reduce patient's waiting time and complaint. Therefore, it is considered that $^{18}F$-Fluoride PET/CT is able to improve diagnosability by offering more accurate images, as cuts in a share of high cost.