Semi-supervised logistic discrimination for functional data

TL;DR

This paper introduces a semi-supervised logistic model for classifying functional data using both labeled and unlabeled datasets, employing EM algorithm for parameter estimation and novel criteria for regularization parameter selection.

Contribution

The paper proposes a new semi-supervised functional logistic model with regularization and model selection criteria, advancing classification methods for functional data.

Findings

Effective in simulations and real data analysis

Improves classification accuracy with unlabeled data

Provides a systematic approach for regularization parameter selection

Abstract

Multi-class classification methods based on both labeled and unlabeled functional data sets are discussed. We present a semi-supervised logistic model for classification in the context of functional data analysis. Unknown parameters in our proposed model are estimated by regularization with the help of EM algorithm. A crucial point in the modeling procedure is the choice of a regularization parameter involved in the semi-supervised functional logistic model. In order to select the adjusted parameter, we introduce model selection criteria from information-theoretic and Bayesian viewpoints. Monte Carlo simulations and a real data analysis are given to examine the effectiveness of our proposed modeling strategy.