Exploratory Analysis of Functional Data via Clustering and Optimal Segmentation

Abstract

We propose in this paper an exploratory analysis algorithm for functional data. The method partitions a set of functions into $K$ clusters and represents each cluster by a simple prototype (e.g., piecewise constant). The total number of segments in the prototypes, $P$, is chosen by the user and optimally distributed among the clusters via two dynamic programming algorithms. The practical relevance of the method is shown on two real world datasets.

References in corpus (5)

• An Algorithm for Optimal Partitioning of Data on an Interval
• Batch and median neural gas
• Fast Selection of Spectral Variables with B-Spline Compression
• A data-driven functional projection approach for the selection of feature ranges in spectra with ICA or cluster analysis
• A regression model with a hidden logistic process for signal parametrization

Cited by in corpus (9)

• Selective review of offline change point detection methods
• Model-based clustering and segmentation of time series with changes in regime
• Application of Visual Clustering Properties of Self Organizing Map in Machine-part Cell Formation
• Model-based functional mixture discriminant analysis with hidden process regression for curve classification
• Neural Networks for Complex Data
• Nonparametric Hierarchical Clustering of Functional Data
• Mixture model-based functional discriminant analysis for curve classification
• Piecewise regression mixture for simultaneous functional data clustering and optimal segmentation
• Unsupervised learning of regression mixture models with unknown number of components