Prediction of Sustainability Status Using Machine Learning Models : The Case of India

Abstract

Sustainable development has emerged as a global imperative, requiring a balanced integration of economic growth, social equity, and environmental protection. India, as one of the world’s most populous and diverse nations, plays a pivotal role in advancing these global objectives. The country has made notable progress in formulating and implementing national policies aligned with the Sustainable Development Goals (SDGs), aiming to reduce socio-economic disparities, enhance infrastructure, and foster environmental sustainability. This study aims to predict the sustainability status across Indian states and union territories using Machine Learning (ML) models. It underscores the transformative potential of data-driven methodologies in evaluating and improving sustainability outcomes. Through the application of classification models (supervised ML) and clustering techniques (unsupervised ML), the study identified key SDG indicators such as electricity access, literacy rate, gender equality, poverty rate, urbanization, healthcare, and malnutrition rate that significantly influence and sustainability. Among the models tested, the Random Forest algorithm achieved the highest predictive accuracy, proving to be a reliable tool for identifying non-sustainable regions and informing targeted policy interventions. Clustering analysis further revealed distinct socio-economic patterns, highlighting the need for region-specific sustainability strategies rather than a one-size-fits-all approach. These findings offer valuable insights for policymakers, urban planners, and sustainability advocates to optimize resource allocation, design strategic interventions, and monitor progress effectively over time.