A hidden process regression model for functional data description. Application to curve discrimination
This work addresses curve discrimination for functional data analysis, particularly in applications like railway monitoring, but it is incremental as it builds on existing regression and hidden process methods.
The authors tackled the problem of modeling functional data with abrupt or smooth regime changes by proposing a regression model with a discrete hidden logistic process, which was applied to curve discrimination. The model was evaluated on simulated and real-world railway switch operation curves, showing competitive performance compared to piecewise regression in modeling and classification tasks.
A new approach for functional data description is proposed in this paper. It consists of a regression model with a discrete hidden logistic process which is adapted for modeling curves with abrupt or smooth regime changes. The model parameters are estimated in a maximum likelihood framework through a dedicated Expectation Maximization (EM) algorithm. From the proposed generative model, a curve discrimination rule is derived using the Maximum A Posteriori rule. The proposed model is evaluated using simulated curves and real world curves acquired during railway switch operations, by performing comparisons with the piecewise regression approach in terms of curve modeling and classification.