Classification and Regression Trees and MLP Neural Network to Classify Water Quality of Canals in Bangkok, Thailand

To evaluate diet quality and relationship between Body Mass Index (BMI), diet quality and inflammatory markers in adolescents of public schools in São Luís-MA. Methodology: A cross-sectional study was conducted with 384 adolescents aged 17 and 18 years. The nutritional status was evaluated through the BMI. The quality of the diet was evaluated through the Revised Diet Quality Index (IQD-R). The inflammatory markers used were C-Reactive Ultrasensitive Protein (hs-CRP), IL-6 (Interleukin-6) and TNF-α (Tumor Necrosis Factor α). Multivariate analysis was performed using a decision tree using the CART (Classification and Regression Trees) algorithm to evaluate the relationship between BMI, diet quality and inflammatory markers. Results: The mean age was 17.3±0.5 years, predominance of females (56.5%) and eutrophic (69.3%). The mean IQD-R score was 55.3±12.7. Adolescents in the lowest tertile of IQD-R (T1) had a higher mean BMI (22.1±4.3 kg/m2 vs 21.5 ± 3.7kg/m2). Higher levels of IL-6 were observed in those located on the IQD-R T1 (1,345 mg/L vs 1,205 mg/L). In the same group (T1), adolescents who had higher IL-6 levels also had a higher mean BMI (23.6±5.1kg/m2 vs 20.8±3.0kg/m2). The adolescents in the largest tertiles of IQD-R (T2 and T3) and who had higher concentrations of IL-6 and CR-us had also a higher mean BMI (23.8±4.9kg/m2). Conclusions: The diet quality of adolescents studied needs modifications. BMI averages varied with diet quality and levels of IL-6 and hs-CRP. Avaliar a qualidade da dieta e a relação entre Índice de Massa Corporal (IMC), qualidade da dieta e marcadores inflamatórios em adolescentes de escolas públicas de São Luís-MA. Metodologia: Foi realizado um estudo transversal com 384 adolescentes de 17 e 18 anos. O estado nutricional foi avaliado por meio do IMC. A qualidade da dieta foi avaliada por meio do Índice de Qualidade da Dieta Revisado (IQD-R). Os marcadores inflamatórios utilizados foram Proteína C Reativa Ultrassensível (PCR-us), IL-6 (Interleucina-6) e TNF-α (Fator de Necrose Tumoral α). A análise multivariada foi realizada usando uma árvore de decisão usando o algoritmo CART (Classification and Regression Trees) para avaliar a relação entre IMC, qualidade da dieta e marcadores inflamatórios. Resultados: A média de idade foi de 17,3 ± 0,5 anos, predomínio do sexo feminino (56,5%) e eutrófico (69,3%). A pontuação média do IQD-R foi de 55,3 ± 12,7. Os adolescentes no tercil inferior do IQD-R (T1) tiveram uma média de IMC mais alta (22,1 ± 4,3kg/m2 vs 21,5 ± 3,7kg/m2). Níveis mais elevados de IL-6 foram observados naqueles localizados no IQD-R T1 (1.345 mg/L vs 1.205 mg/L). No mesmo grupo (T1), os adolescentes que apresentaram níveis mais elevados de IL-6 também apresentaram média de IMC mais elevada (23,6 ± 5,1kg/m2 vs 20,8 ± 3,0kg/m2). Os adolescentes nos maiores tercis de IQD-R (T2 e T3) e que apresentaram maiores concentrações de IL-6 e CR-us também apresentaram maior IMC médio (23,8 ± 4,9kg/m2). Conclusões: A qualidade da dieta dos adolescentes estudados necessita de modificações. As médias do IMC variaram com a qualidade da dieta e os níveis de IL-6 e PCR-us.

Download Full-text

Komparasi Akurasi Algoritme CART Dan Neural Network Untuk Diagnosis Penyakit Diabetes Retinopathy

CogITo Smart Journal ◽

10.31154/cogito.v7i1.304.121-134 ◽

2021 ◽

Vol 7 (1) ◽

pp. 121

Author(s):

Pungkas Subarkah ◽

Muhammad Marshal Abdallah ◽

Septi Oktaviani Nur Hidayah

Keyword(s):

Neural Network ◽

Data Mining ◽

Cross Validation ◽

Confusion Matrix ◽

Regression Trees ◽

Classification And Regression Trees ◽

Diabetes Retinopathy ◽

Classification And Regression ◽

F Measure ◽

Diabetes Melitus

Penyakit Diabetes Retinopathy atau DR adalah salah satu komplikasi mikrovaskular diabetes melitus dengan angka prevalensi yang cukup tinggi yang bisa menyebabkan kematian. Penderita DR hingga saat ini masih sulit disembuhkan karena mayoritas penderita melakukan pemeriksaan di saat kondisi penyakit telah memasuki tahap berbahaya, hal ini dikarenakan sifat dari penyakit DR ini tidak menunjukkan gejala yang terlihat bila masih pada tahap awal. Penelitian ini menguji diagnosis penyakit diabetes retinopathy dengan melakukan klasiifikasi menggunakan metode data mining. Metode yang digunakan ialah algoritme Classification And Regression Trees (CART) dan Algoritme Neural Network menggunakan dataset diambil dari UCI Repository Learning diperoleh daro Universitas Debreen, Hongaria. Adapun metode validasi dan evaluasi yang digunakan dalam penelitian ini yaitu 10-cross validation dan confusion matrix. Hasil dari akurasi pada algoritme CART yaitu 63.4231% dengan nilai precision 0.64%, Recall 0.634%, dan F-Measure 0.634% dan algoritme Neural Network mendapatkankan nilai akurasi sebesar 72.285% dengan nilai precision 0.723%, Recall 0.723%, dan F-Measure 0.723%. Dari hasil tersebut dapat disimpulkan bahwa algoritme Neural Network lebih baik dalam mendiagnosis penyakit diabetes retinopathy. Kata kunci— Klasifikasi, Diagnosis, Diabetes Retinopathy, Algoritme, CART, Neural Network

Download Full-text

Application and Comparison of Neural Network, C5.0, and Classification and Regression Trees (CART) algorithms in the Credit Risk Evaluation Problem (Case Study: A Standard German Credit Dataset)

International Journal of Knowledge Engineering and Data Mining ◽

10.1504/ijkedm.2017.10006181 ◽

2017 ◽

Vol 4 (1) ◽

pp. 1

Author(s):

Fahimeh Kheradmand ◽

Alireza Arshadi Khamseh ◽

Mahdi Massahi Khoraskani,

Keyword(s):

Neural Network ◽

Credit Risk ◽

Risk Evaluation ◽

Regression Trees ◽

Classification And Regression Trees ◽

Problem Case ◽

Evaluation Problem ◽

Classification And Regression

Download Full-text

The feature extraction of water-quality data by using classification and regression trees

BUNSEKI KAGAKU ◽

10.2116/bunsekikagaku.52.331 ◽

2003 ◽

Vol 52 (5) ◽

pp. 331-336

Author(s):

Mikio KAIHARA ◽

Nobuyuki TAKAHASHI ◽

Seikou SATO ◽

Yohko HIGUCHI ◽

Satomi YUSA

Keyword(s):

Water Quality ◽

Feature Extraction ◽

Regression Trees ◽

Classification And Regression Trees ◽

Quality Data ◽

Water Quality Data ◽

Classification And Regression

Download Full-text

Preoperative hematocrit and platelet count are associated with blood loss during spinal fusion for children with neuromuscular scoliosis

Journal of Perioperative Practice ◽

10.1177/1750458920962634 ◽

2021 ◽

pp. 175045892096263

Author(s):

Margaret O Lewen ◽

Jay Berry ◽

Connor Johnson ◽

Rachael Grace ◽

Laurie Glader ◽

...

Keyword(s):

Platelet Count ◽

Blood Loss ◽

Spinal Fusion ◽

Regression Trees ◽

Classification And Regression Trees ◽

Neuromuscular Scoliosis ◽

Independent Variables ◽

Estimated Blood Loss ◽

Laboratory Results ◽

Classification And Regression

Aim To assess the relationship of preoperative hematology laboratory results with intraoperative estimated blood loss and transfusion volumes during posterior spinal fusion for pediatric neuromuscular scoliosis. Methods Retrospective chart review of 179 children with neuromuscular scoliosis undergoing spinal fusion at a tertiary children’s hospital between 2012 and 2017. The main outcome measure was estimated blood loss. Secondary outcomes were volumes of packed red blood cells, fresh frozen plasma, and platelets transfused intraoperatively. Independent variables were preoperative blood counts, coagulation studies, and demographic and surgical characteristics. Relationships between estimated blood loss, transfusion volumes, and independent variables were assessed using bivariable analyses. Classification and Regression Trees were used to identify variables most strongly correlated with outcomes. Results In bivariable analyses, increased estimated blood loss was significantly associated with higher preoperative hematocrit and lower preoperative platelet count but not with abnormal coagulation studies. Preoperative laboratory results were not associated with intraoperative transfusion volumes. In Classification and Regression Trees analysis, binary splits associated with the largest increase in estimated blood loss were hematocrit ≥44% vs. <44% and platelets ≥308 vs. <308 × 109/L. Conclusions Preoperative blood counts may identify patients at risk of increased bleeding, though do not predict intraoperative transfusion requirements. Abnormal coagulation studies often prompted preoperative intervention but were not associated with increased intraoperative bleeding or transfusion needs.

Download Full-text

Regional Mapping of Groundwater Potential in Ar Rub Al Khali, Arabian Peninsula Using the Classification and Regression Trees Model

Remote Sensing ◽

10.3390/rs13122300 ◽

2021 ◽

Vol 13 (12) ◽

pp. 2300

Author(s):

Samy Elmahdy ◽

Tarig Ali ◽

Mohamed Mohamed

Keyword(s):

Machine Learning ◽

Regional Scale ◽

Regression Trees ◽

Classification And Regression Trees ◽

Groundwater Potential ◽

Machine Learning Algorithms ◽

Conditioning Factors ◽

Potential Mapping ◽

Classification And Regression ◽

Groundwater Potential Mapping

Mapping of groundwater potential in remote arid and semi-arid regions underneath sand sheets over a very regional scale is a challenge and requires an accurate classifier. The Classification and Regression Trees (CART) model is a robust machine learning classifier used in groundwater potential mapping over a very regional scale. Ten essential groundwater conditioning factors (GWCFs) were constructed using remote sensing data. The spatial relationship between these conditioning factors and the observed groundwater wells locations was optimized and identified by using the chi-square method. A total of 185 groundwater well locations were randomly divided into 129 (70%) for training the model and 56 (30%) for validation. The model was applied for groundwater potential mapping by using optimal parameters values for additive trees were 186, the value for the learning rate was 0.1, and the maximum size of the tree was five. The validation result demonstrated that the area under the curve (AUC) of the CART was 0.920, which represents a predictive accuracy of 92%. The resulting map demonstrated that the depressions of Mondafan, Khujaymah and Wajid Mutaridah depression and the southern gulf salt basin (SGSB) near Saudi Arabia, Oman and the United Arab Emirates (UAE) borders reserve fresh fossil groundwater as indicated from the observed lakes and recovered paleolakes. The proposed model and the new maps are effective at enhancing the mapping of groundwater potential over a very regional scale obtained using machine learning algorithms, which are used rarely in the literature and can be applied to the Sahara and the Kalahari Desert.

Download Full-text