<p>Acidophiles and alkaliphiles are microorganisms that thrive in extremely acidic and alkaline environments, respectively. Although extensive research has been conducted on the classification of thermophilic enzymes, far fewer studies have focused on the classification of acidic enzymes, particularly using machine learning. This study aims to classify acidic and alkaline enzymes within the α/β hydrolase family using machine learning methods based on their amino acid sequences and physicochemical properties, independent of secondary structural variations. In this study, 403 bacterial enzymes from the α/β hydrolase family were analyzed to classify acidic and alkaline enzymes. We applied several machine learning models, including Decision Tree, Random Forest, eXtreme Gradient Boosting, and Support Vector Machine, along with analyses of amino acid sequences and their chemical properties. The Random Forest model achieved 76% accuracy and an AUC of 0.90 ± 0.03, revealing that alanine, lysine, and the grand average of hydropathicity (GRAVY) were key features contributing to the classification of acidic and alkaline enzymes. Alanine and glycine showed a positive correlation with acidic characteristics, suggesting their importance in acidic environments and their potential as targets for enzyme engineering. The identified key amino acids and physicochemical properties provide insights into enzyme engineering and industrial applications under extreme conditions.</p>