A Review on Machine Learning and Deep Learning based Rainfall Prediction Methods

Water for farming and storage capacity in dams are both affected by the quantity of precipitation that falls. Predicting when and how much rain will fall is challenging because of global warming and other variables. Artificial Intelligence (AI), Machine Learning (ML), and Deep Learning (DL) may use factors like the amount of rainfall in previous years, climatic change, climatic conditions, and so on to forecast rainfall in a given area. Potential flooding and damage to man-made structures might result from unexpected rainfall. Comprehensive data categorization and investigation, evaluation, inquiry, and interpretations are needed to uncover rainfall that will help governments avert human deaths, agricultural failures, and animal fatalities. The development of ML and DL systems is ongoing to assess and manage crucial rainfall data. Therefore, the purpose of this research is to investigate the potential of using ML and DL approaches to forecast precipitation for future planning of human settlements and related activities. Prediction methods in ML are investigated, including regression, Convolutional Neural Networks (CNNs), and Long-Short Term Memory (LSTM). The benefits and drawbacks of using neural network algorithms and ML algorithms are also discussed in this paper, with each approach taking into account the data and features at hand. New developments in the field of precipitation forecasting are also covered, including studies that use a wide range of feature extraction and prediction methods. In this study, we discuss many ML methods and algorithms for predicting rainfall based on historical data and other criteria.