Weather Forecasting using Machine Learning for Smart Farming

Weather forecast is of prime attention of the researchers working in the smart agriculture domain. In India, approximately 55% of the total crops are dependent on weather (monsoon season). An accurate weather forecast model requires abundant data to get the most accurate predictions. However, the weather forecast is a key area of research and is always challenging from historical data. Hence, the current system used for weather forecasting is an amalgamation of forecasting models, opinions, and information trends, and specific patterns. This work presents the application of the linear regression model and polynomial regression model for weather forecasting; like a scheme to forecast rainfall, and precipitation using historical weather data. The sample weather dataset covers 75 districts of Uttar Pradesh state which is received from the Indian Meteorological Department (IMD). Furthermore, analysing the impact of forecasts with different parameters is realized over six major crops Triticum (biological name of wheat), Gram, Barley, Mustard, Sugarcane, and Maize of Uttar Pradesh State. The main objective of the state-of-the-art is efficient crop management and passing the appropriate message to farmers to make suitable decisions as per the weather conditions.