A Bayesian Lasso Functional Clustering Model

• Activity Number: 460 - Clustering Methods for Big Data Problems
• Type: Topic Contributed
• Date/Time: Wednesday, August 2, 2017 : 8:30 AM to 10:20 AM
• Sponsor: Section on Statistical Learning and Data Science
• Author(s): Alejandro Murua* and Folly Adjogou and Wolfgang Raffelsberger
• Companies: Universite de Montreal and Université de Montréal and Institut de Génetique et de Biologie Moléculaire et Cellulaire, Université de Strasbourg
• Keywords: functional data analysis ; clustering ; classification ; lasso penalty ; gene expression data

Abstract

We introduce a Bayesian lasso penalty model for time-course data. The goal is to do clustering of functional data. Our model combines functional data analysis with model based clustering. The functional framework is used to model time-course data. Principal functional components are described by score coefficients which embed the curves in a much lower dimensional space. Model based clustering is performed on the score space, thus avoiding the curse of dimensionality in the curves space. The number of clusters as well as the dimension of the score space are determined via a Bayesian lasso penalty model. Monte Carlo techniques are used in order to estimate the normalizing constants for different values of the penalty parameters. One of the advantages of the Bayesian lasso model is that it avoids the need to perform a costly cross-validation to select the penalty parameters. We present an application to the analysis of gene-expression data associated with the effects of exposure to tobacco smoke.