funLOCI: A Local Clustering Algorithm for Functional Data

Article Statistics & Probability

Probabilistic K-means with Local Alignment for Clustering and Motif Discovery in Functional Data

Marzia A. A. Cremona et al.

Summary: We propose a new method for locally clustering curves and discovering functional motifs, which are typical shapes that may recur multiple times along the curves capturing important local characteristics. Our method combines ideas from functional data analysis, bioinformatics, and fuzzy clustering to identify shared curve portions. It can incorporate various dissimilarity measures and derivatives to exploit complex facets of shapes. We demonstrate its performance with simulation studies and real data applications to Covid-19 death curves and mutagenesis-related Omics data.

JOURNAL OF COMPUTATIONAL AND GRAPHICAL STATISTICS (2023)

Add to Collection

Article Computer Science, Interdisciplinary Applications

FunCC: A new bi-clustering algorithm for functional data with misalignment

Marta Galvani et al.

Summary: The paper addresses the problem of bi-clustering functional data, introducing a novel method called FunCC, which can handle data misalignment and jointly cluster and align curves. Simulation studies and a real case study demonstrate the potential and advantages of the method.

COMPUTATIONAL STATISTICS & DATA ANALYSIS (2021)

Add to Collection

Article Statistics & Probability