An investigation of how normalisation and local modelling techniques confound machine learning performance in a mental health study

Keyphrases

• Fuzzy Scoring 100%
• Machine Learning Performance 100%
• Quantile Normalization 50%
• Local Model 50%
• Heterogeneous Data 25%
• Parameter Setting 25%
• ROC-AUC 25%
• Development Features 25%
• Technology Classification 25%
• Biomedical Studies 25%
• Health Datasets 25%
• Data Processing Methods 25%
• Naivete 25%
• Clinical Mental Health 25%
• Data Heterogeneity 25%
• Prognostic Technology 25%
• Combinatorial Reasoning 25%
• Goods Classification 25%

Computer Science

• Cluster Formation 33%
• Data Quality Issue 33%
• Data Heterogeneity 33%
• Processing Method 33%
• Biomedical Study 33%
• Data Modeling 33%
• Strong Influence 33%
• Classification Performance 33%
• Heterogeneous Data 33%

Biochemistry, Genetics and Molecular Biology

• Mental Health 100%
• Feature Extraction 100%