Classification with Feature Selection and Extraction Methods
-
Updated
Aug 30, 2021 - Python
Classification with Feature Selection and Extraction Methods
A Python implementation of feature selection algorithms using k-Nearest Neighbor classification. This project implements three different search strategies for finding optimal feature subsets: Forward Selection, Backward Elimination, and Simulated Annealing.
Classes reproducing certain results from sklearn
Uses nearest neighbor algorithm to find which feature is the best indicator for a certain class attribute
Used CDC dataset for heart attack detection in patients. Balanced the dataset using SMOTE and Borderline SMOTE and used feature selection and machine learning to create different models and compared them based on metrics such as F-1 score, ROC AUC, MCC, and accuracy.
Self-Validated Ensemble Models (SVEM) for Python: lasso/elastic-net and forward-selection ensembles for small-sample design of experiments. Python counterpart of the R package SVEMnet.
To associate your repository with the forward-selection topic, visit your repo's landing page and select "manage topics."