Classification of fertilizer type based on soil minerals using voting classification over k-nearest neighbour algorithm

• Published in: AIP conference proceedings Vol. 2822; no. 1
• Main Authors: Bandaiah, K., Parvathy, L. Rama
• Format: Journal Article Conference Proceeding
• Language: English
• Published: Melville American Institute of Physics 14.11.2023
• Subjects: Accuracy; Algorithms; Classifiers; Fertilizers; Machine learning; Minerals; Soil classification; Soils
• ISSN: 0094-243X; 1551-7616
• DOI: 10.1063/5.0172896

Using the vote classifier, predict the type of fertiliser based on soil minerals. For forecasting the accuracy % of fertiliser type, a Voting Classifier with a sample size of 10 and a K-Nearest Neighbor (KNN) with a sample size of 10 were iterated at different times. A Voting Classifier is a machine learning model that trains on a large ensemble of models and predicts an output (class) based on the highest likelihood of the chosen class being the outcome. The findings shown that Voting Classifier achieved substantial results with 96% accuracy when compared to KNN with 96% accuracy. The voting classifier and KNN have a statistical significance of p=0.001 (p<0.05). The most successful algorithm for classifying fertiliser types based on soil minerals is the Voting Classifier than KNN.