Repo for documentation and testing codebase for no-code platform with omop cdm
One project the repo includes is advanced feature selection. Feature selection is a critical step in machine learning to improve model performance, reduce overfitting, and enhance interpretability. The advanced feature selection project extracts candidate predictors from omop and submits the candidates to one or more feature selection methods like the fisher score and forward selection. Fisher score is a filter method that ranks features based on the ratio of between-class variance to within-class variance. Forward selection is a so-called "wrapper method" that starts with no features and adds one feature at a time that improves model performance. See here for a landscape analysis of feature selection methods
This project can be part of a larger one where we first employ feature extraction followed by feature selection followed by some kind of handoff to the No Code ML platform.