Classification of Fertiliser Type Based on Soil Minerals Using Voting Classification Over Decision Tree

Abstract

To estimate the type of fertilizer based on the soil minerals using the voting classifier. For forecasting fertilizer type accuracy %, a Voting Classifier with a sample size of 10 and a Decision Tree with a sample size of 10 was iterated at various times. A supervised learning algorithm is a Decision Tree. It constructs a “forest” using an array of decision trees, typically trained to use the “bagging” method. A Novel Voting Classification is a predictive model that learns from several models and predicts an output (class) based on the result representing the greatest likelihood of being the chosen class. The Novel Voting Classifier produced substantial results with 96 percent accuracy, compared to 94% accuracy for the Decision Tree. The Novel voting classifier and the Decision Tree showed statistical evidence of p=0.001 (p<0.05). Voting Classifier is the most effective algorithm that classifies the type of fertilizer based on soil minerals with more accuracy than the Decision Tree.