• International Journal of Computer Science & Network Security
• /
• 제23권8호
• /
• pp.17-25
• /
• 2023

The alarming global prevalence of Type 2 Diabetes Mellitus (T2DM) has catalyzed an urgent need for robust, early diagnostic methodologies. This study unveils a pioneering approach to predicting T2DM, employing the Extreme Gradient Boosting (XGBoost) algorithm, renowned for its predictive accuracy and computational efficiency. The investigation harnesses a meticulously curated dataset of 4303 samples, extracted from a comprehensive Chinese research study, scrupulously aligned with the World Health Organization's indicators and standards. The dataset encapsulates a multifaceted spectrum of clinical, demographic, and lifestyle attributes. Through an intricate process of hyperparameter optimization, the XGBoost model exhibited an unparalleled best score, elucidating a distinctive combination of parameters such as a learning rate of 0.1, max depth of 3, 150 estimators, and specific colsample strategies. The model's validation accuracy of 0.957, coupled with a sensitivity of 0.9898 and specificity of 0.8897, underlines its robustness in classifying T2DM. A detailed analysis of the confusion matrix further substantiated the model's diagnostic prowess, with an F1-score of 0.9308, illustrating its balanced performance in true positive and negative classifications. The precision and recall metrics provided nuanced insights into the model's ability to minimize false predictions, thereby enhancing its clinical applicability. The research findings not only underline the remarkable efficacy of XGBoost in T2DM prediction but also contribute to the burgeoning field of machine learning applications in personalized healthcare. By elucidating a novel paradigm that accentuates the synergistic integration of multifaceted clinical parameters, this study fosters a promising avenue for precise early detection, risk stratification, and patient-centric intervention in diabetes care. The research serves as a beacon, inspiring further exploration and innovation in leveraging advanced analytical techniques for transformative impacts on predictive diagnostics and chronic disease management.

• Journal of Preventive Medicine and Public Health
• /
• 제56권1호
• /
• pp.41-49
• /
• 2023

Objectives: Rwanda reported a stunting rate of 33% in 2020, decreasing from 38% in 2015; however, stunting remains an issue. Globally, child deaths from malnutrition stand at 45%. The best options for the early detection and treatment of stunting should be made a community policy priority, and health services remain an issue. Hence, this research aimed to develop a model for predicting stunting in Rwandan children. Methods: The Rwanda Demographic and Health Survey 2019-2020 was used as secondary data. Stratified 10-fold cross-validation was used, and different machine learning classifiers were trained to predict stunting status. The prediction models were compared using different metrics, and the best model was chosen. Results: The best model was developed with the gradient boosting classifier algorithm, with a training accuracy of 80.49% based on the performance indicators of several models. Based on a confusion matrix, the test accuracy, sensitivity, specificity, and F1 were calculated, yielding the model's ability to classify stunting cases correctly at 79.33%, identify stunted children accurately at 72.51%, and categorize non-stunted children correctly at 94.49%, with an area under the curve of 0.89. The model found that the mother's height, television, the child's age, province, mother's education, birth weight, and childbirth size were the most important predictors of stunting status. Conclusions: Therefore, machine-learning techniques may be used in Rwanda to construct an accurate model that can detect the early stages of stunting and offer the best predictive attributes to help prevent and control stunting in under five Rwandan children.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2021년도 추계학술발표대회
• /
• pp.681-684
• /
• 2021

In this paper, we study image clustering without labeling using machine learning techniques. We proposed an unsupervised machine learning technique to design an image clustering model that automatically categorizes images into groups. Our experiment focused on inception convolutional neural networks (inception V3) with k-mean methods to cluster images. For this, we collect the public datasets containing Food-K5, Flowers, Handwritten Digit, Cats-dogs, and our dataset Rice Germination, and the owner dataset Palm print. Our experiment can expand into three-part; First, format all the images to un-label and move to whole datasets. Second, load dataset into the inception V3 extraction image features and transferred to the k-mean cluster group hold on six classes. Lastly, evaluate modeling accuracy using the confusion matrix base on precision, recall, F1 to analyze. In this our methods, we can get the results as 1) Handwritten Digit (precision = 1.000, recall = 1.000, F1 = 1.00), 2) Food-K5 (precision = 0.975, recall = 0.945, F1 = 0.96), 3) Palm print (precision = 1.000, recall = 0.999, F1 = 1.00), 4) Cats-dogs (precision = 0.997, recall = 0.475, F1 = 0.64), 5) Flowers (precision = 0.610, recall = 0.982, F1 = 0.75), and our dataset 6) Rice Germination (precision = 0.997, recall = 0.943, F1 = 0.97). Our experiment showed that modeling could get an accuracy rate of 0.8908; the outcomes state that the proposed model is strongest enough to differentiate the different images and classify them into clusters.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2022년도 추계학술발표대회
• /
• pp.478-480
• /
• 2022

This paper studies human activity image classification using deep transfer learning techniques focused on the inception convolutional neural networks (InceptionV3) model. For this, we used UFC-101 public datasets containing a group of students' behaviors in mathematics classrooms at a school in Thailand. The video dataset contains Play Sitar, Tai Chi, Walking with Dog, and Student Study (our dataset) classes. The experiment was conducted in three phases. First, it extracts an image frame from the video, and a tag is labeled on the frame. Second, it loads the dataset into the inception V3 with transfer learning for image classification of four classes. Lastly, we evaluate the model's accuracy using precision, recall, F1-Score, and confusion matrix. The outcomes of the classifications for the public and our dataset are 1) Play Sitar (precision = 1.0, recall = 1.0, F1 = 1.0), 2), Tai Chi (precision = 1.0, recall = 1.0, F1 = 1.0), 3) Walking with Dog (precision = 1.0, recall = 1.0, F1 = 1.0), and 4) Student Study (precision = 1.0, recall = 1.0, F1 = 1.0), respectively. The results show that the overall accuracy of the classification rate is 100% which states the model is more powerful for learning UCF-101 and our dataset with higher accuracy.

• 한국지하수토양환경학회지:지하수토양환경
• /
• 제29권1호
• /
• pp.39-50
• /
• 2024

Slags from steel-making industry have been recycled at a target rate of 95% and most of them are recycled as media-contact type such as fill and cover materials in Korea. However, as they contain free phase CaO during their generation, they may not only expand and collapse upon contact with water, but high pH leachate and heavy metals leaching may occur. In this study, the Korean leaching procedure (KLP) and up-flow percolation test were performed for the samples collected from 17 steel-making production plants in Korea. The waste quality criteria were met in all tests, but pH of the samples was above 10. There are no regulations on the pH of leachate in most of the countries, however, Germany, Italy, and Australia have set a pH range of 10 to 13 for the leachates. Although slag leachate cannot be considered hazardous based only on its high pH, it is necessary to reduce the pH of leachate to minimize the impact on the surrounding environment. Furthermore, conflicting regulations on wastes handling and management in Korea created confusion on the types of wastes subject to recycling. Therefore, an appropriate management plan for steel-making slags needs to be established. To this end, this study attempted to provide a guideline for managing steel-making slag waste by considering international guidelines and current management practices in Korea.

• 한국수자원학회논문집
• /
• 제57권3호
• /
• pp.181-193
• /
• 2024

기후변화에 의해 발생하는 대기 불균형은 강우량의 증가로 이어지고, 침수 발생 빈도가 증가함에 따라 이를 탐지할 수 있는 기술의 필요성이 증가하고 있다. 침수 피해를 최소화하기 위해 지속적인 모니터링이 필요하며, 날씨의 영향을 받지 않는 합성개구레이더(Synthetic Aperture Radar, SAR) 영상을 활용하여 침수지역을 탐지하였다. 관측된 데이터는 median 필터를 통해 노이즈를 감소시키는 전처리 과정을 진행하였으며, 객체 탐지 기법을 통해 수체와 비수체를 분류하여 각 기법의 침수탐지 활용성을 평가하고자 하였다. 본 연구에서는 Otsu 기법과 SVM 기법을 통해 수체 및 침수 탐지를 수행하였으며, Confusion Matrix를 통해 전체적인 모델의 성능을 평가하였다. Otsu 기법은 수체와 비수체의 경계를 구분하는데 적합함을 보였으나, 혼합물의 영향을 받아 오탐지의 비율이 높게 나타났다. 반면, SVM 기법을 사용한 경우, 오탐지 비율이 낮고 혼합물에 의한 영향에 민감하지 않은 것으로 관측되었다. 이에 따라 침수 상태를 제외한 다른 조건에서 SVM 기법의 정확도가 높게 나타났다. Otsu 기법이 침수 조건에서 SVM 기법보다 다소 높은 정확도를 보였지만, 정확도의 차이가 5% 미만임을 확인할 수 있었다(Otsu: 0.93, SVM: 0.90). SVM 기법이 Otsu 기법보다 침수 전, 침수 후의 조건에서 정확도 차이가 최대 15% 이상 발생하여 수체 및 침수탐지에 더 적합하게 나타났다(Otsu: 0.77, SVM: 0.92). 이러한 결과는 SVM 기법이 수체 및 침수탐지에서 효과적으로 활용될 수 있음을 시사하며, 미래의 수재해 탐지 시스템에 적용될 때 유용한 정보를 제공할 수 있을 것으로 기대된다.

• Tuberculosis and Respiratory Diseases
• /
• 제74권5호
• /
• pp.207-214
• /
• 2013

Background: Community-acquired pneumonia (CAP) is one of the leading causes of death among the elderly. Several studies have reported the clinical usefulness of serum procalcitonin, a biomarker of bacterial infection. However, the association between the levels of procalcitonin and the severity in the elderly with CAP has not yet been reported. The aim of this study was to evaluate usefulness of procalcitonin as a predictor of severity and mortality in the elderly with CAP. Methods: This study covers 155 CAP cases admitted to Pusan National University Hospital between January 2010 and December 2010. Patients were divided into two groups (${\geq}65$ years, n=99; <65 years, n=56) and were measured for procalcitonin, C-reactive protein (CRP), white blood cell, confusion, uremia, respiratory rate, blood pressure, 65 years or older (CURB-65) and pneumonia severity of index (PSI). Results: The levels of procalcitonin were significantly correlated with the CURB-65, PSI in totals. Especially stronger correlation was observed between the levels of procalcitonin and CURB-65 in the elderly (procalcitonin and CURB-65, ${\rho}$=0.408 with p<0.001; procalcitonin and PSI, ${\rho}$=0.293 with p=0.003; procalcitonin and mortality, ${\rho}$=0.229 with p=0.023). The correlation between the levels of CRP or WBC and CAP severity was low. The existing cut-off value of procalcitonin was correlated with mortality rate, however, it was not correlated with mortality within the elderly. Conclusion: The levels of procalcitonin are more useful than the levels of CRP or WBC to predict the severity of CAP. However, there was no association between the levels of procalcitonin and mortality in the elderly.

• 한국산림과학회지
• /
• 제102권2호
• /
• pp.176-181
• /
• 2013

산지계류에서 계상구조물의 설치에 따른 저서성 대형무척추동물의 출현종 및 개체수의 변화에 따른 서식처 영향을 분석한 결과를 요약하면 다음과 같다. 계상구조물의 종류 및 위치에 따른 저서성 대형무척추동물의 출현종은 최소 16종에서 최대 40종 범위이며, 개체수는 최소 352 $inds./^2$에서 최대 4,333.3 $inds./m^2$ 범위로 나타났다. 돌바닥막이, 돌낙차공 및 콘크리트바닥막이와 같은 계상구조물의 설치는 계상의 안정을 주어 종적인 침식을 막고 유속을 감소시키는 역할을 하지만 구조물 인접지점과 상류부 및 하류부에 수심이 깊고 단조로운 소와 폭호가 형성되어 있는 지점은 계류의 균일한 구조적인 연속성을 저해하고 미소서식처의 인위적 및 자연적인 교란으로 서식공간이 제한되어 저서성 대형무척추동물의 출현종수 및 개체수가 줄어 생물 다양성이 감소되는 것을 알 수 있다. 따라서, 산지계류의 생태적 안정을 위하여 계획 단계에서부터 생물서식처의 훼손이 최소화될 수 있도록 계상구조물이 설치될 위치 주변에 서식공간의 확보 등 적절한 대책이 필요할 것으로 판단된다.

• Environmental Engineering Research
• /
• 제13권3호
• /
• pp.125-130
• /
• 2008

Free ammonia ($NH_3$-N) inhibition of nitrite-oxidizing bacteria (NOB) has been widely studied for partial nitrification (or nitrite accumulation) and denitrification via nitrite ($NO_2^-$-N) as a low-cost treatment of ammonium containing wastewater. The literature on $NH_3$-N inhibition of NOB, however, shows disagreement about the threshold $NH_3$-N concentration and its degree of inhibition. In order to clarify the confusion, a simple and cheap respirometric method was devised to investigate the effect of free ammonia inhibition of NOB. Sludge samples from an autotrophic nitrifying reactor were exposed to various $NH_3$-N concentrations to measure the maximum specific nitrite oxidation rate ($\hat{K}_{NO}$) using a respirometer. NOB biomass was estimated from the yield values in the literature. Free ammonia inhibition of nitrite oxidizing bacteria was reversible and the specific nitrite oxidation rate ($K_{NO}$) decreased from 0.141 to 0.116, 0.100, 0.097 and 0.081 mg $NO_2^-$-N/mg NOB h, respectively, as the $NH_3$-N concentration increased from 0.0 to 1.0, 4.1, 9.7 and 22.9 mg/L. A nonlinear regression based on the noncompetitive inhibition mode gave an estimate of the Inhibition concentration ($K_I$) of free ammonia to be 21.3 mg $NH_3$-N/L. Previous studies gave $\hat{K}_{NO}$ of Nitrobacter and Nitrospira as 0.120 and 0.032 mg/mg VSS h. The free ammonia concentration which inhibits Nitrobacter was $30{\sim}50\;mg$ $NH_3$-N/L and Nitrospira was inhibited at $0.04{\sim}0.08\;mg$ $NH_3$-N/L. The results support the fact that Nitrobacter is the dominant NOB in the reactor. The variations in the reported values of free ammonia inhibition may be due to the different species of nitrite oxidizers present in the reactors. The respirometric method provides rapid and reliable analysis of the behavior and community of the nitrite oxidizing bacteria.

• 한국수학교육학회지시리즈C:초등수학교육
• /
• 제27권3호
• /
• pp.303-320
• /
• 2024

본 연구는 폴리아의 문제해결 단계에 따라 풀이를 제공하는 GPTs 기반 맞춤형 챗봇을 제작하여 수학적 성능을 검증하였다. 우선 맞춤형 챗봇 베타 버전을 제작하여 수학적 성능을 검증한 후 대표적인 오류를 수정하여 최종 맞춤형 챗봇을 완성하였다. 완성된 맞춤형 챗봇은 초등 수학 6학년 교과서에 제시된 이미지 형태의 65개 문제 중 평균 약 57.8개를 옳게 해결하여 약 89.0%의 정답률을 보였으며, 베타 버전에 비해 약 4%p 높은 정답률을 나타냈다. 또한 그림이 문제를 해결하는 데 핵심적인 역할을 하지 않는 50개 문제의 경우 평균 45.5개를 옳게 해결하여 약 91.0%의 정답률을 보였다. 완성된 맞춤형 챗봇의 답변 중 대표적인 오류는 문제 인식 오류이며, 문제에 인식하기 어려운 그림이 사용되었거나 문제 구성이 복잡한 경우에 해당 오류가 나타났다. 다음으로 개념 혼동 오류, 문제 이해 오류 등이 나타났다. 본 연구에서 개발한 문제해결 맞춤형 챗봇은 범용적인 챗봇인 ChatGPT보다 우수한 수학적 성능을 보였다. 또한 학년 수준에 적절하도록 풀이 과정의 조정이 가능하여 학생 개별화 맞춤형 지도에 활용할 수 있으며, 누구나 제작이 가능하여 교사 개인별 수업 보조 등 수학교육에서의 다양한 활용 가능성을 엿볼 수 있다.