Feature Engineering and Model Selection

This chapter focuses on feature engineering and model selection in machine learning microbiome data. In addition to the Summary, there are four sections. Section 6.1 introduces general feature engineering, including a feature engineering definition, goals, and methods. Section 6.2 introduces model selection in statistics and machine learning, including definitions of model selection, variable selection, and statistical modeling, and details of model selection and variable selection in statistics and machine learning. Section 6.3 reviews feature engineering in microbiome research and particularly focuses on three newly developed feature engineering methods and provides some remarks on hierarchical feature engineering: (1) hierarchical feature engineering, (2) microbiome preprocessing machine learning pipeline, and (3) the TaxaHFE algorithm. Section 6.4 introduces and illustrates how to perform feature selection using the Kruskal-Wallis rank-sum test combined with the functions simper_pretty() and krusk_pretty() and provides remarks on feature selection using the Kruskal-Wallis rank-sum test.