Comparative performances of machine learning methods for classifying Crohn Disease patients using genome-wide genotyping data

Keyphrases

• Machine Learning Techniques 100%
• Crohn's Disease 100%
• Gradient Boosted Trees 100%
• Area under the ROC Curve 100%
• Genome-wide Genotyping 100%
• Logistic Regression 66%
• Genome-wide Association Study 66%
• Genetic Architecture 66%
• Quality Control Method 66%
• Genetic Variants 33%
• Healthy Controls 33%
• Nonlinear Model 33%
• Artificial Neural Network 33%
• XGBoost 33%
• Genetic Markers 33%
• Immunochip 33%
• Neural Network 33%
• Inflammatory Bowel Disease Genetics 33%
• Patient-controlled 33%
• Hidden Layer 33%
• Genetic Disease 33%
• Regression Results 33%
• Machine Learning Approach 33%
• Encoding Method 33%
• Penalized Logistic Regression 33%
• Least Absolute Shrinkage and Selection Operator (LASSO) 33%
• Elastic Net 33%
• Best Predictor 33%
• Tree Network 33%
• Complementary Approach 33%
• LightGBM 33%
• Missing Genotypes 33%
• Tree-based Methods 33%
• Logistic Regression Method 33%
• Maximal Area 33%
• Epistatic Effect 33%
• Genomic Information 33%
• Marker Preselection 33%
• Machine Settings 33%
• CatBoost 33%
• Dense Neural Network 33%
• Encoding Strategy 33%
• Quality Coding 33%

Biochemistry, Genetics and Molecular Biology

• Genotyping 100%
• Genome-Wide Association Study 100%
• Genetic Architecture 100%
• Genetic Disorder 100%
• Quality Control 100%
• Genetic Divergence 50%
• Genetic Marker 50%
• Artificial Neural Network 50%
• Genomics 50%