A feature optimization module for high-dimensional datasets (bioinformatics-leaning). It automates variance filtering and backward elimination to remove redundant features while preserving meaningful predictive signal. Designed to boost classifier performance, interpretability, and efficiency.

Project exploring the impact of node pruning on CNN performance metrics within a stratified k-fold cross-validation framework. Focuses on optimizing recall for medical imaging tasks while analyzing trade-offs in precision and accuracy