Prediction of Seed Germination Quality Utilizing Ensemble-Based Precision Forming

Abstract

Seed germination is a primary objective of precision agriculture. Precision agriculture, which makes extensive use of machine learning, has been the subject of recent studies on predictive analytics. These machine learning methods typically employ supervised learning models to make predictions about how successfully seeds will germinate. However, a major challenge that modern models face when attempting to make accurate predictions is the curse of dimensionality in the training corpus. The primary contribution of this manuscript is an ensemble-based method for predicting seed germination quality (EL-GQP) in precision agriculture. The accuracy of predictions can be improved using this ensemble method, which combines the positive aspects of a number of different models while minimising the negative aspects of the individual models. The proposed model is significantly superior to the current model, as demonstrated by experimental results of cross-validation on the benchmark dataset. During the simulation, work is done on the corpus dataset contains 4250 negative records and 6230 positive records.