Description

Farmers and agronomists now employ sensors to aid with operational improvement. To remotely monitor their crops, they use sensor data sent over IoT. In the name of modern farming, farmers today manipulate the environment in which their crops are grown to maximize yields. The harshness of the weather and changes in disease, on the other hand, have an impact on crop productivity. This paper’s main goal is to introduce the Multisensor Machine-Learning Approach (MMLA), a revolutionary method for classifying multisensor data. The fusion technique supports high-quality data analysis for cultivation suggestions in agricultural environments. Eight crops were categorized based on the suggested recommendation system: cotton, gram, groundnut, maize, moong, paddy, sugarcane, and wheat. Three machine learning methods were used to categorize different types of crops:Decision Tree, KNN for finding the suitable crop for the season. In this project take temperature, humidity as a feature for finding the crop which is suitable for the climate and also calculate the accuracy and classification report and also calculate the confusion matrix for the prediction of the crop.