Likelihood of Transformation to Green Infrastructure Using Ensemble Machine Learning Techniques in Jinan, China

Rapid urbanization influences green infrastructure (GI) development in cities. The government plans to optimize GI urban areas, which requires understanding spatiotemporal trends areas and driving forces influencing their pattern. Traditional GIS-based methods, used determine the greening potential of vacant land are incapable predicting future scenarios based on past trend. Therefore, we propose a heterogeneous ensemble technique spatial pattern Jinan, China, biophysical socioeconomic factors. Data-driven artificial neural networks (ANN) random forests (RF) selected as base learners, while support vector machine (SVM) is meta classifier. Results showed that stacking model ANN-RF-SVM achieved best test accuracy (AUC 0.941) compared individual ANN, RF, SVM algorithms. Land surface temperature, distance water bodies, population density, rainfall found be most factors regarding conversion Jinan.