Internet of Things plays an important role in agriculture monitoring area. The Weather prediction supports to the farmers make prediction in rainfall season through prior identification. Farmers want help maximizing boom efficiency, protecting assets and optimizing manufacturing through IoT. Changing climate patterns have implications for all walks of existence. Shrewd and complex weather forecasting is important to early forecast to reduce the impact of climate styles. Agriculture is a huge industry that is laid low with weather change. To acquire those desires, farmers need a weather forecasting result for planting and irrigation. Rainfall prediction using Internet of Things (IOT) sensors is a tough challenge in weather forecasting. Problem in machine earning techniques have feature hidden patterns in recorded weather records. Rainfall forecast refers back to the understanding of weather condition parameters, together with temperature, air strain, humidity, wind pace, etc. Rainfall forecast is the priority for early prediction of rainfall IOT sensors which helps each farmers and the people. Because the general public in India depend on agriculture. This paper introduces the Linear Regression Rainfall Prediction Technique (LRRPT) for rainfall estimation to select the importance of features. Often there may be a range at the same time to analyze the occurrence of weather conditions with weather impact rate (WIR). The system performs a forecasting process based on historical weather information and produces Categories of rainfall, along with wind velocity and temperature. In keeping with the characteristic importance the ANFIS neural network is carried out to expect the rainfall forecasting depends on the climate records carried out to test the forecasting technique. Experiments show that the system can achieve better prediction accuracy 91% precision 93% recall 94% and f-measure 92% in rainfall forecast. © 2024 Elsevier B.V., All rights reserved.