Analyzing Public Environmental Awareness Using Advanced Machine Learning for Sustainable Urban Transportation
Sustainable Development, vol.33, no.6, pp.8619-8637, 2025 (SSCI, Scopus)
- Publication Type: Article / Article
- Volume: 33 Issue: 6
- Publication Date: 2025
- Doi Number: 10.1002/sd.70122
- Journal Name: Sustainable Development
- Journal Indexes: Social Sciences Citation Index (SSCI), Scopus, IBZ Online, International Bibliography of Social Sciences, PASCAL, ABI/INFORM, Agricultural & Environmental Science Database, Aqualine, Aquatic Science & Fisheries Abstracts (ASFA), Business Source Elite, Business Source Premier, CAB Abstracts, Environment Index, Geobase, Greenfile, Index Islamicus, PAIS International, Political Science Complete, Pollution Abstracts, Sociological abstracts, Veterinary Science Database, Worldwide Political Science Abstracts, Civil Engineering Abstracts
- Page Numbers: pp.8619-8637
- Keywords: environmental awareness, explainable artificial intelligence, generative adversarial networks, machine learning, sustainable transportation
- Erciyes University Affiliated: Yes
Abstract
Environmental awareness and sustainable transportation are critical in addressing climate change and urbanization. This research enhances understanding of environmental awareness through advanced machine learning (ML), focusing on cycling as a key component of sustainable urban mobility. Based on survey data from 550 participants in Kayseri, Türkiye, the study examines demographic, behavioral, and attitudinal factors influencing environmental awareness. Bioinspired feature selection algorithms, including genetic algorithm and particle swarm optimization, identified key predictors. Generative Adversarial Networks (GANs) generated synthetic data for underrepresented groups, improving dataset balance and reliability. Seven classification models were evaluated using 10-fold cross-validation. Ensemble methods, particularly CatBoost and LightGBM, achieved over 0.82 accuracy with balanced precision, recall, and F1-score. Behavioral factors, such as reasons for choosing a bicycle and environmental expectations, were the most significant determinants. These findings can inform targeted cycling infrastructure planning, inclusive environmental campaigns, and the development of predictive tools to identify vulnerable or responsive user groups.