Rainfall forecasting is crucial for several sectors in India, including agriculture, water resource management, determining the causes of ground subsidence that has affected Joshimath (Uttarakhand), India and disaster relief. This study forecasts the patterns of rainfall in several regions of India using machine learning techniques. The research uses historical meteorological data on temperature, humidity, wind speed, and other relevant variables to develop prediction models. When it comes to rainfall prediction, the precision of several machine learning approaches is evaluated, including decision tree regression, Random Forest, MLR, XGBoost, AdaBoost, Bagging, KNN, and Linear Discriminant Analysis (LDA). The results demonstrate how machine learning may increase rainfall prediction accuracy, providing crucial new knowledge for risk mitigation and sustainable development in India.