A Novel Feature Selection Technique to Better Predict Climate Change Stage of Change

Indications of people’s environmental concern are linked to transport decisions and can provide great support for policymaking on climate change. This study aims better predict individual change stage (CC-SoC) based different features transport-related behavior, General Ecological Behavior, New Environmental Paradigm, socio-demographic characteristics. Together these sources result in over 100 possible that indicate someone’s level concern. Such a large number may create several analytical problems, such as overfitting, accuracy reduction, high computational costs. To this end, new feature selection technique, named the Coyote Optimization Algorithm-Quadratic Discriminant Analysis (COA-QDA), is first proposed find optimal CC-SoC with highest accuracy. Different conventional methods (Lasso, Elastic Net, Random Forest Feature Selection, Extra Trees, Principal Component Selection) employed compare COA-QDA. Afterward, eight classification techniques applied solve prediction problem. Finally, sensitivity analysis performed determine most important affecting CC-SoC. The results COA-QDA outperforms by increasing average testing data from 0.7% 5.6%. Logistic Regression surpasses other classifiers