Classifying Environmental Attitudes Using Explainable Machine Learning Models: A Study on Interpretability in Environmental Psychology

The importance of environmental attitudes and environmental education is examined in this research. Negative impressions can worsen ecological difficulties, therefore knowing public attitudes is crucial as environmental issues grow. To uncover elements that impact environmental views, this study classifies and analyzes them using machine learning. Environmental awareness and behavior survey results are used in the research. With 94% accuracy, Logistic Regression was the best machine learning model tested. The model's accuracy shows its capacity to identify the environmental attitudes of people. Explainable AI improved Logistic Regression model interpretability. This technique reveals the model's decision-making process and the most important element of attitude categorization. The most important element in environmental views was the response to “I do not pay attention to whether the products I buy have a recycling symbol on their packaging.” These findings imply that consumer awareness of product packaging and recycling symbols influences environmental sustainability views. This study has several ramifications. The findings highlight the need for focused environmental education programs that promote recycling and sustainable consumption. Educational programs may engage people and promote beneficial environmental behavior by addressing the attitudes highlighted in this study.