Table 1 Classification accuracy of both Decision Tree and Logistic Regression on dataset with no feature selection and feature selection using FCBF and FCBFK algorithms.

From: A Rapid Response Electrochemical Biosensor for Detecting Thc In Saliva

  Raw Impedance Data (no normalization) Normalized to Synthetic Saliva Measurement Normalized to Antibody Measurement
Decision Tree Logistic Regression Decision Tree Logistic Regression Decision Tree Logistic Regression
No Feature Selection 0.710 0.713 0.746 0.721 0.765 0.724
FCBF 0.901 0.845 0.824 0.813 0.699 0.632
FCBFK 0.820 0.702 0.652 0.813 0.742 0.746