Classification of EEG recordings in auditory brain activity via a logistic functional linear regression model

TL;DR

This paper introduces a logistic functional linear regression model using wavelet representations to classify EEG recordings for early auditory phonemic categorization, exploring various penalized likelihood and reduction techniques.

Contribution

It presents a novel approach combining wavelet-based functional data analysis with logistic regression for EEG classification, comparing multiple reduction methods.

Findings

Wavelet representation effectively captures EEG features.

Penalized likelihood improves classification accuracy.

Principal component and PLS reductions enhance model performance.

Abstract

We want to analyse EEG recordings in order to investigate the phonemic categorization at a very early stage of auditory processing. This problem can be modelled by a supervised classification of functional data. Discrimination is explored via a logistic functional linear model, using a wavelet representation of the data. Different procedures are investigated, based on penalized likelihood and principal component reduction or partial least squares reduction.