Optimal Variable Selections and Prediction Methods for Soil Chemical Property Using Data Mining and Machine Learning Approaches

Abstract

The main objective of the research is to convert raw data into actionable knowledge. Creating a decision tree involves recursively dividing the data into subsets based on different attributes, aiming to achieve homogeneity or minimize variance within each subset. This paper considers agriculture and its soil chemical-related dataset for applying data mining techniques to find suitable variables for future predictions. The tenfamiliar ML approaches are Gaussian processes,linear regression, multilayer perceptron, simple linear regression, SMOreg, decision stump, M5P, random forest, random tree, and REP tree. Numerical illustrations are provided to prove the proposed results with test statistics or accuracy parameters.