• Journal of the Korean Data and Information Science Society
• /
• v.26 no.1
• /
• pp.255-260
• /
• 2015

Among many optimal threshold criteria from ROC curve, the closest-to-(0,1) and amended closest-to-(0,1) criteria are considered. An ROC curve that passes close to the (0,1) point indicates that two models are well classified. In this case, the ROC curve is located far from the (1,0) point. Hence we propose two criteria: the farthest-to-(1,0) and amended farthest-to-(1,0) criteria. These criteria are found to have a relationship with the KolmogorovSmirnov statistic as well as some optimal threshold criteria. Moreover, we derive that a definition for the proposed criteria with more than two dimensions and with relations to multi-dimensional optimal threshold criteria.

• Journal of Environmental Science International
• /
• v.18 no.7
• /
• pp.767-780
• /
• 2009

This study investigates the standard, duration period and excess mortality of extreme heat using the standardized daily mortality data from 1991 to 2004, establishing a standard threshold Criteria for Heat Health Watch Warning System in Korea. It ultimately aims to provide the guidance in building up Heat Health Watch Warning System for Korea by suggesting the standard to quantify thermal stress from heat. The standard threshold Criteria for Heat Health Watch Warning System for Seoul metropolitan city takes into account both daily maximum temperature and daily maximum heat index(HI) and consists of four phases; caution, extreme caution, danger, and extreme danger. Extreme caution phase and danger phase are used as the advisory and warning of extreme heat, respectively. Since the nationwide distribution of the frequency of extreme heat day and the excess mortality rate shows little difference across regions, the standard threshold Criteria for Heat Health Watch Warning System for Seoul metropolitan city can be used for other regions.

• The Korean Journal of Applied Statistics
• /
• v.27 no.5
• /
• pp.773-786
• /
• 2014

We propose the multiplication of false rates (MFR) which is a classification accuracy criteria and an area type of rectangle from ROC curve. Optimal threshold obtained using MFR is compared with other criteria in terms of classification performance. Their optimal thresholds for various distribution functions are also found; consequently, some properties and advantages of MFR are discussed by comparing FNR and FPR corresponding to optimal thresholds. Based on general cost function, cost ratios of optimal thresholds are computed using various classification criteria. The cost ratios for cost curves are observed so that the advantages of MFR are explored. Furthermore, the de nition of MFR is extended to multi-dimensional ROC analysis and the relations of classification criteria are also discussed.

• Proceedings of the Computational Structural Engineering Institute Conference
• /
• 2007.04a
• /
• pp.751-756
• /
• 2007

In this paper, perception threshold for horizontal vibration of tall buildings was investigated. After a comparative study of human comfort criteria for wind-induced vibration in foreign countries being made, perception threshold was recorded by increasing acceleration in the range of 0.2Hz through 1.2Hz of frequency in horizontal vibration experiments, and perception of subjects was examined by a proper questionnaire. Also, the results obtained from experiments of horizontal vibration were compared with Japanese standard(AIJES-A001-2004).

• Restorative Dentistry and Endodontics
• /
• v.41 no.3
• /
• pp.159-166
• /
• 2016

The purpose of this review was to suggest practical criteria for the clinical translucency evaluation of direct esthetic restorative materials, and to review the translucency with these criteria. For the evaluation of reported translucency values, measuring instrument and method, specimen thickness, background color, and illumination should be scrutinized. Translucency parameter (TP) of 15 to 19 could be regarded as the translucency of 1 mm thick human enamel. Visual perceptibility threshold for translucency difference in contrast ratio (${\Delta}CR$) of 0.07 could be transformed into ${\Delta}TP$ value of 2. Translucency differences between direct and indirect resin composites were perceivable (${\Delta}TP>2$). Universal and corresponding flowable resin composites did not show perceivable translucency differences in most products. Translucency differed significantly by the product within each shade group, and by the shade group within each product. Translucency of human enamel and perceptibility threshold for translucency difference may be used as criteria for the clinical evaluation of translucency of esthetic restorative materials.

• Journal of Korean Institute of Industrial Engineers
• /
• v.17 no.2
• /
• pp.107-113
• /
• 1991

This article develops a control model for an open queueing network in terms of both the input and the output processes with stochastic intensities. The input and the output intensities are subject to some capacity limits and optimum control is characterized by a threshold type with a finite upper barrier. A discounted profit is used as a decision criteria, which is revenue minus operating and holding cost.

• Journal of Environmental Science International
• /
• v.17 no.1
• /
• pp.57-66
• /
• 2008

The frequency analysis and the probability plot were applied to heavy metal contents of soils collected from the Goseong Cu mine area as a statistic method for the determination of the threshold value which was able to partition a population comprising largely dispersed heavy metal contents into the background and the anomalous populations. Almost all the heavy metal contents of soil showed a positively skewed distributions and their cumulative percentage frequencies plotted as a curved lines on logarithmic probability plot which represent a mixture of two or more overlapping populations. Total Cu, Pb and Cd data and extractable Cu and Pb data could be partitioned into background and anomalous populations by using the inflection in each curve. The others showed a normally distributed population or an largely overlapped populations. The threshold values obtained from replotted frequency distributions with the partitioned populations were Cu 400 mg/kg, Pb 450 mg/kg and Cd 3.5 mg/kg in total contents and Cu 40 mg/kg and Pb 12 mg/kg in extractable contents, respectively. The thresholds for total contents are much higher than the tolerable level of soil pollution proposed by Kloke(Cu 100 mg/kg, Pb 100 mg/kg, Cd 3 mg/kg), but those for extractable contents are not exceeded the worrying level of soil pollution proposed by Ministry of Environment(Cu 50 mg/kg, Pb 100 mg/kg). When the threshold values were used as the criteria of soil pollution in the study area, $9{\sim}19%$ of investigated soil population was in polluted level. The spatial distributions of heavy metal contents greater than threshold values showed that polluted soils with heavy metals are restricted within the mountain soils in the vicinity of abandoned mines.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.14 no.5
• /
• pp.1181-1185
• /
• 1990

A failure evaluation diagram to evaluate fatigue fracture was developed. The relation between the fatigue limit and the threshold stress intensity factor for the short-cracked specimens of various materials including a piping carbon steel can be rationally predicted by the proposed method. It is shown that the coupled failure evaluation diagram for fatigue and ductile fracture is expecially useful for evaluation of the flaw tolerance as well as the margin of the safety of the pressure vessel and piping. Further, accumulation of fatigue data will be needed to construct an accurate fatigue failure evaluation diagram.

• Journal of the Korean Data and Information Science Society
• /
• v.25 no.6
• /
• pp.1521-1530
• /
• 2014

The ROC analysis is considered for multiple class diagnosis. There exist many criteria to find optimal thresholds and measure the accuracy of diagnostic tests for k dimensional ROC analysis. In this paper, we proposed a diagnostic accuracy measure called the correct classification simple rate, which is defined as the summation of true rates for each classification distribution and expressed as a function of summation of sequential true rates for two consecutive distributions. This measure does not weight accuracy across categories by the category prevalence and is comparable across populations for multiple class diagnosis. It is found that this accuracy measure does not only have a relationship with Kolmogorov - Smirnov statistics, but also can be represented as a linear function of some optimal threshold criteria. With these facts, the suggested measure could be applied to test for comparing multiple distributions.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.18 no.2
• /
• pp.59-65
• /
• 2023

Currently, polysomnography is essential to diagnose sleep-related breathing disorders. However, there are several disadvantages to polysomnography, such as the requirement for multiple sensors and a long reading time. In this paper, we propose a system for predicting the severity of sleep-related breathing disorders at home utilizing measurable elements in a wearable device. To predict severity, the variables were refined through a three-step variable selection process, and the refined variables were used as inputs into three machine-learning models. As a result of the study, random forest models showed excellent prediction performance throughout. The best performance of the model in terms of F1 scores for the three threshold criteria of 5, 15, and 30 classified as the AHI index was about 87.3%, 90.7%, and 90.8%, respectively, and the maximum performance of the model for the three threshold criteria classified as the RDI index was approx 79.8%, 90.2%, and 90.1%, respectively.