Rainfall Prediction using Machine Learning Techniques – A Comparative Approach

India depends heavily on agriculture for its survival. Rainfall is crucial to agriculture. Predicting rainfall has become a significant issue recently. People are made aware of the possibility of rain and are better prepared to protect their crops by taking preventative measures when rain is predicted. Making an accurate rainfall prediction has always been challenging, but in recent years, because of climatic variability and change, it has become even more challenging. The effectiveness of classification systems for predicting rainfall has soared. For the purpose of predicting rainfall in the Indian states, this research employs a variety of categorization techniques. The classification algorithms are Logistic Regression by Lesso, Decision Tree, Light GBM, Extreme Gradient Boosting, Random Forest Method, K Nearest Neighbor Technique, Naive Bayes and Support Vector Machine Techniques. Dataset, which included several meteorological parameters, was obtained via Kaggle. With a testing data ratio of 70:30. Based on precision, accuracy, recall, f1-score, and execution time, the efficacy of the classification algorithms was assessed. There are few challenges with the existing system with data and with accuracy which are the major problems in predictions and to understand which algorithm is best especially for rainfall prediction our aim is to find measures for all possible algorithms and provide a conclusion saying the algorithm that fits for rainfall prediction.