Predictive Models for Rice Quality Traits Using MATLAB

The purpose of this dataset is to provide predictive models for assessing rice quality traits (e.g., amylose content, viscosity profiles) and classifying rice types using Near-Infrared (NIR) spectroscopy. These models were developed in MATLAB (Matlab R2023a) using advanced methods, including Partial Least Squares (PLS), Partial Least Squares Discriminant Analysis (PLS-DA), Support Vector Machines (SVM), Machine Learning, and Artificial Neural Networks (ANN), applied to NIR spectra data. The PLS, iPLS, siPLS and mwPLS models were performed using MATLAB software and the iToolbox for MATLAB available from (http://www.models.life.ku.dk/itoolbox). The Classification toolbox (version 2.0), developed by Milano Chemometrics and QSAR Research Group (http//michem.disat.unimib.it/chm), was used for classification procedure. This dataset includes the following components: 1. AMYLOSE_MODEL_FOOD_CHEMISTRY A MATLAB-based predictive model for determining rice amylose content using NIR spectroscopy combined with PLS chemometric algorithms. 2. PLS-DA_SVM_MODELS MATLAB-based predictive models for identifying different rice flour types using NIR spectroscopy with PLS-DA and SVM methods. 3. ANN_MODELS_ApSci MATLAB-based predictive models for determining rice pasting parameters using NIR spectroscopy and Machine Learning tools. 4. ANN_MODELS_PASTING_FOODS A MATLAB-based predictive model for evaluating rice quality based on grain physical parameters using Artificial Neural Network methods. The .mat files are provided in a compressed .rar format and include the models with model performance metrics, ensuring reproducibility and facilitating model evaluation.