Variable selection and updating in model-based discriminant analysis for high dimensional data with food authenticity applications

TL;DR

This paper introduces a semi-supervised, variable selection discriminant analysis method tailored for high-dimensional food authenticity data, demonstrating superior classification accuracy and interpretability over existing methods.

Contribution

The paper presents a novel semi-supervised discriminant analysis approach with an efficient variable selection strategy for high-dimensional food authenticity data.

Findings

Achieved excellent classification performance on multiple food authenticity datasets.

Selected meaningful variables that enhance interpretability.

Outperformed standard machine learning methods like Random Forests and SVMs.

Abstract

Food authenticity studies are concerned with determining if food samples have been correctly labeled or not. Discriminant analysis methods are an integral part of the methodology for food authentication. Motivated by food authenticity applications, a model-based discriminant analysis method that includes variable selection is presented. The discriminant analysis model is fitted in a semi-supervised manner using both labeled and unlabeled data. The method is shown to give excellent classification performance on several high-dimensional multiclass food authenticity data sets with more variables than observations. The variables selected by the proposed method provide information about which variables are meaningful for classification purposes. A headlong search strategy for variable selection is shown to be efficient in terms of computation and achieves excellent classification performance. In applications to several food authenticity data sets, our proposed method outperformed default implementations of Random Forests, AdaBoost, transductive SVMs and Bayesian Multinomial Regression by substantial margins.