funOCLUST: Clustering Functional Data with Outliers

TL;DR

funOCLUST introduces a robust clustering method for functional data that effectively handles outliers, extending the OCLUST algorithm to infinite-dimensional settings and demonstrating strong performance on various datasets.

Contribution

It extends the OCLUST algorithm to functional data, providing a robust clustering approach that manages outliers in infinite-dimensional spaces.

Findings

Effective clustering of functional data

Strong outlier detection capabilities

Validated on simulated and real datasets

Abstract

Functional data present unique challenges for clustering due to their infinite-dimensional nature and potential sensitivity to outliers. An extension of the OCLUST algorithm to the functional setting is proposed to address these issues. The approach leverages the OCLUST framework, creating a robust method to cluster curves and trim outliers. The methodology is evaluated on both simulated and real-world functional datasets, demonstrating strong performance in clustering and outlier identification.