Machine Learning Algorithms for Protein Physicochemical Component Prediction Using Near Infrared Spectroscopy in Chickpea Germplasm

Abstract

Prediction of physicochemical components of chickpea flour using near infrared spectroscopy requires discovering exact wavelength regions that provide the most useful data before preprocessing. This study used six essential machine learning techniques to develop models for predicting proteinphysicochemical component in chickpea: Linear Regression (LR), Artificial Neural Network (ANN), Partial Least Squares Regression (PLSR), Random Forest (RF), Support Vector Regression (SVR) and Decision Tree Regression (DTR). Performance measurements such as Root Mean Square Error and Karl Pearson’s Correlation Coefficient and Coefficient of Determination were used to validate the models. RF and ANN models showed significant improvement over all other models in terms of accuracy.