Internet of Things Based Weekly Crop Pest Prediction by Using Deep Neural Network

Internet of Things (IoT) assisted application in agriculture shows tremendous success to improve productivity agriculture. Agriculture is grappling with issues such as depleted soil fertility, climate-related hazards like intensified pest attacks and diseases. Accurate forecasting outbreaks can play a vital role improving agricultural yield. Utilizing IoT technology for environmental monitoring crop fields forecast attacks. The important parameters predictions are temperature, humidity, rainfall, wind speed sunshine duration. Directly sensed conditions utilized input deep learning model, which makes binary decisions about the presence populations based on prevailing conditions. accuracy precision model making assessed through evaluation test data. Five year data 2028 2022 have been used prediction. prediction generates weekly predictions. overall 94% high F-measure, Precision, Recall, Cohens kappa, ROC AUC optimize improves gradually time. Weekly generated from means all environment last seven days. short-term measures against