Efficient and sparse feature selection for biomedical text classification via the elastic net: Application to ICU risk stratification from nursing notes

Keyphrases

• Risk Stratification 100%
• Intensive Care Unit 100%
• Elastic Net 100%
• Sparse Feature Selection 100%
• Unit Risk 100%
• Efficient Features 100%
• Nursing Notes 100%
• Biomedical Data Classification 100%
• Sparsity 66%
• Sparse Model 66%
• Intensive Care Unit Patients 66%
• Log Loss 66%
• Sparse Representation Classifier 66%
• Intensive Care Unit Mortality 66%
• Clinically Significant 33%
• Statistical Model 33%
• Loss Function 33%
• Impact Performance 33%
• Mortality Prediction 33%
• Intensive Care Unit Admission 33%
• Mortality Risk 33%
• Nonclinical 33%
• Regularization Parameter 33%
• Model Interpretation 33%
• Number of Features 33%
• Feature Selection Methods 33%
• Free Text 33%
• Mortality Model 33%
• Hinge Loss 33%
• Elastic Net Regularization 33%
• L1 Regularization 33%
• Stochastic Gradient Descent 33%
• L2 Regularization 33%
• Adult Intensive Care Unit 33%
• Text Features 33%
• Feature Select 33%
• MIMIC-III Database 33%

Computer Science

• Feature Selection 100%
• Net Application 100%
• Text Classification 100%
• Regularization 100%
• Feature Extraction 100%
• Sparsity 66%
• Mortality Risk 66%
• Desirable Property 33%
• Regularization Parameter 33%
• Stochastic Gradient Descent 33%