• Journal of the Korean Data and Information Science Society
• /
• v.25 no.6
• /
• pp.1263-1272
• /
• 2014

Longitudinal categorical data commonly occur from medical, health, and social sciences. In these data, the correlation of repeated outcomes is taken into account to explain the effects of covariates exactly. In this paper, we introduce marginalized random effects models that are used for the estimation of the population-averaged effects of covariates. We also review how these models have been developed. Real data analysis is presented using the marginalized random effects.

• Journal of Korean Society of Rural Planning
• /
• v.11 no.3 s.28
• /
• pp.11-17
• /
• 2005

Land prices reflect not only the uses of land, but the potential uses as well(Plantinga, 2002) so land values can be applied to very effective indices for deciding regional status and growing potential. The purpose of this study is to deduce determinant factors of regional land prices. Principal determinants of regional land prices are analyzed with a hedonic technique and spatial econometric models based on 2001 statistic data of Korea except large cities. The results provide the followings. 1. The spatial effect of rural regions are very little with adjacent regions. 2. The common index of land price is population density and other determinant factors are different depending on land uses.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.9
• /
• pp.2904-2926
• /
• 2022

Currently, diabetes is the most common chronic disease in the world, affecting 23.7% of the population in the Kingdom of Saudi Arabia. Diabetes may be the cause of lower-limb amputations, kidney failure and blindness among adults. Therefore, diagnosing the disease in its early stages is essential in order to save human lives. With the revolution in technology, Artificial Intelligence (AI) could play a central role in the early prediction of diabetes by employing Machine Learning (ML) technology. In this paper, we developed a diagnosis system using machine learning models for the detection of type 2 diabetes among adults, through the adoption of two different diabetes datasets: one for training and the other for the testing, to analyze and enhance the prediction accuracy. This work offers an enhanced classification accuracy as a result of employing several pre-processing methods before applying the ML models. According to the obtained results, the implemented Random Forest (RF) classifier offers the best classification accuracy with a classification score of 98.95%.

• Asian Journal of Atmospheric Environment
• /
• v.11 no.3
• /
• pp.194-201
• /
• 2017

This study investigates the factors that affect China's air pollution using city-level panel data and spatial econometric models. We address three air pollutants ($PM_{10}$, $SO_2$, and $NO_2$) present in 30 cities in China between 2004-2012 using global OLS and spatial models. To develop the spatial econometric analysis, we create a spatial weights matrix to define spatial patterns based on two neighborhood criteria - the queen contiguity and k nearest neighbors. The results show that the estimated coefficients are relatively consistent across different spatial weight criteria. The OLS models indicate that the effect of green spaces is statistically significant in decreasing the concentrations of all air pollutants. In the $PM_{10}$ and $SO_2$ analyses, the OLS models find that the number of buses and population density are also positively related to a reduction in the concentration of air pollutants. In addition, an increase in the temperature and the presence of secondary industries increase $SO_2$ and $NO_2$ concentrations, respectively. All spatial models capture a positive and significant effect of green spaces on reducing the concentration of each air pollutant. Our results suggest that green spaces in cities should receive priority consideration in local planning aimed at sustainable development. Furthermore, policymakers need to be able to discern the differences among pollutants when establishing environmental policies.

• Asian Pacific Journal of Cancer Prevention
• /
• v.15 no.10
• /
• pp.4143-4145
• /
• 2014

Background: This study used receiver operating characteristic curve to analyze Surveillance, Epidemiology and End Results (SEER) Ewing sarcoma (ES) outcome data. The aim of this study was to identify and optimize ES-specific survival prediction models and sources of survival disparities. Materials and Methods: This study analyzed socio-economic, staging and treatment factors available in the SEER database for ES. 1844 patients diagnosed between 1973-2009 were used for this study. For the risk modeling, each factor was fitted by a Generalized Linear Model to predict the outcome (bone and joint specific death, yes/no). The area under the receiver operating characteristic curve (ROC) was computed. Similar strata were combined to construct the most parsimonious models. Results: The mean follow up time (S.D.) was 74.48 (89.66) months. 36% of the patients were female. The mean (S.D.) age was 18.7 (12) years. The SEER staging has the highest ROC (S.D.) area of 0.616 (0.032) among the factors tested. We simplified the 4-layered risk levels (local, regional, distant, un-staged) to a simpler non-metastatic (I and II) versus metastatic (III) versus un-staged model. The ROC area (S.D.) of the 3-tiered model was 0.612 (0.008). Several other biologic factors were also predictive of ES-specific survival, but not the socio-economic factors tested here. Conclusions: ROC analysis measured and optimized the performance of ES survival prediction models. Optimized models will provide a more efficient way to stratify patients for clinical trials.

• Journal of Korean Public Health Nursing
• /
• v.31 no.1
• /
• pp.5-18
• /
• 2017

Purpose: This study was conducted to suggest a method for financial projection of health insurance expenditures that reflects future changes in demographic structure. Methods: Using data associated with the number of patients and health insurance cost per patient, generalized linear models (GLM) were fitted with demographic explanatory variables. Models were constructed separately for individual medical departments, types of medical service, and types of public health insurance. Goodness-of-fit of most of the applied GLM models was quite satisfactory. By combining estimates of frequency and severity from the constructed models and results of the population projection, total annual health insurance expenditures were projected through year 2060. Results: Expenditures for medical departments associated with diseases that are more frequent in elderly peoples are expected to increase steeply, leading to considerable increases in overall health insurance expenditures. The suggested method can contribute to improvement of the accuracy of financial projection. Conclusion: The overall demands for medical service, medical personnel, and relevant facilities in the future are expected to increase as the proportion of elderly people increases. Application of a more reasonable estimation method reflecting changes in demographic structure will help develop health policies relevant to above mentioned resources.

• Journal of Forest and Environmental Science
• /
• v.28 no.3
• /
• pp.158-169
• /
• 2012

Sabah has experienced a rapid decline in the extent of forest cover. The precise impact of habitat loss on the conservation status of the plants of Sabah is uncertain. In this study we use the niche modelling algorithm MAXENT to construct preliminary, revised and final ecological niche models for Dipterocarpus lamellatus and Dipterocarpus ochraceus and combined these models with data on current land-use to derive conservation assessments for each species. Preliminary models were based on herbarium data alone. Ground surveys were conducted to evaluate the performance of these preliminary models, and a revised niche model was generated from the combined herbarium and ground survey data. The final model was obtained by constraining the predictions of the revised models by filters. The range overlap between the preliminary and revised models was 0.47 for D. lamellatus and 0.39 for D. ochraceus, suggesting poor agreement between them. There was substantial variation in estimates of habitat loss for D. ochraceus, among the preliminary, revised and constrained models, and this has the potential to lead to incorrect threat assessments. From these estimates of habitat loss, the historic distribution and estimates of population size we determine that both species should be classified as Critically Endangered under IUCN Red List guidelines. Our results suggest that ground-truthing of ecological niche models is essential, especially if the models are being used for conservation decision making.

• Asian-Australasian Journal of Animal Sciences
• /
• v.24 no.12
• /
• pp.1644-1650
• /
• 2011

This study was conducted to detect quantitative trait loci (QTL) affecting meat quality in an $F_2$ reference population of Korean native pig and Landrace crossbreds. The three-generation mapping population was generated with 411 progeny from 38 $F_2$ full-sib families, and 133 genetic markers were used to produce a sex-average map of the 17 autosomes. The data set was analyzed using least squares Mendelian and parent-of-origin interval-mapping models. Lack-of-fit tests between models were used to characterize the QTL for mode of gene expressions. A total of 10 (32) QTL were detected at the 5% genome (chromosome)-wise level for the analyzed traits. Of the 42 QTL detected, 13 QTL were classified as Mendelian, 10 as paternal, 14 as maternal, and 5 as partial expressed QTL, respectively. Among the QTL detected at 5% genome-wise level, four QTL had Mendelian mode of inheritance on SSCs 5, 10, 12, and 13 for cooking loss, drip loss, crude lipid and crude protein, respectively; two QTL maternal inheritance for pH at 24-h and shear force on SSC11; three QTL paternal inheritance for CIE b and Hunter b on SSC9 and for cooking loss on SSC15; and one QTL partial expression for crude ash on SSC13, respectively. Most of the Mendelian QTL (9 of 13) had a dominant mode of gene action, suggesting potential utilization of heterosis for genetic improvement of meat quality within the cross population via marker-assisted selection.

• The Bulletin of The Korean Astronomical Society
• /
• v.38 no.2
• /
• pp.56.1-56.1
• /
• 2013

Supernovae type Ia (SNe Ia) cosmology is providing the only direct evidence for the presence of dark energy. This result is based on the assumption that the look-back time evolution of SNe Ia luminosity, after light-curve shape correction, would be negligible. However, the most recent compilation of SNe Ia data shows systematic difference in the Hubble residual (HR) between the E and Sd/Irr galaxies, indicating that the light-curve fitters used by the SNe Ia community cannot quite correct for a large portion of the population age effect. In order to investigate this possibility more directly, we have obtained low-resolution spectra for 30 nearby early-type host galaxies. This data set is used to estimate the luminosity-weighted mean ages and metallicities of host galaxies by employing the population synthesis models. We found an interesting trend between the host galaxy age and HR, in the sense that younger galaxies have positive residuals (i.e., light-curve corrected SNe Ia luminosity is fainter). This result is rather independent of the choice of the population synthesis models employed. Taken at face value, this age (evolution) effect can mimic a large fraction of the HR used in the discovery of the dark energy. This result is significant at 1.4 - 3 sigma levels, depending on the light curve fitters adopted, and further observations and analyses are certainly required to confirm the trend reported here.

• Molecular & Cellular Toxicology
• /
• v.4 no.2
• /
• pp.124-131
• /
• 2008

The growing problem of obesity is associated with numerous medical conditions. Several studies have reported that activation of thyroid hormone receptor beta $(THR{\beta})$ is involved in lipid metabolism and thermogenesis. To identify the relationship between the $THR{\beta}$ gene and obesity, we genotyped eighty two single nucleotide polymorphisms (SNPs) in the gene using the Affymetrix array chip in 209 overweight/obese and 155 normal subjects in Korean population. Of the eighty two polymorphisms, the seven SNPs exhibited a significant association with overweight/obesity in three alternative models (codominant, dominant, and recessive models; P<0.05 after adjusting for age and sex) were rs826221 (+267878 T>C), rs4858604 (+186399 A>G), rs1158265 (+200152 T>C), rs1868575 (+206031 G>A), rs1700939 (+238467 T>A), rs1505301 (+241933 T>C), and rs1924768 (+126491 T>C). During haplotype analysis using HapAnalyzer software, 2 haplotypes (block 13: TTAT; block 15: CTGC) containing significant polymorphisms (rs1700939 +238467 T>A and rs4858604 +186399 A>G) were detected to be significantly different. The results suggest that the $THR{\beta}$ gene may be associated with overweight/obesity in Korean population.