Projection pursuit based on Gaussian mixtures and evolutionary algorithms

TL;DR

This paper introduces a novel projection pursuit method using Gaussian mixture models and genetic algorithms to identify informative low-dimensional projections of multivariate data for visualization.

Contribution

It presents a semi-parametric approach combining GMMs and GAs for flexible, effective projection pursuit, enhancing data visualization and structure detection.

Findings

Effective on artificial datasets

Demonstrated on real datasets

Improves visualization of complex structures

Abstract

We propose a projection pursuit (PP) algorithm based on Gaussian mixture models (GMMs). The negentropy obtained from a multivariate density estimated by GMMs is adopted as the PP index to be maximised. For a fixed dimension of the projection subspace, the GMM-based density estimation is projected onto that subspace, where an approximation of the negentropy for Gaussian mixtures is computed. Then, Genetic Algorithms (GAs) are used to find the optimal, orthogonal projection basis by maximising the former approximation. We show that this semi-parametric approach to PP is flexible and allows highly informative structures to be detected, by projecting multivariate datasets onto a subspace, where the data can be feasibly visualised. The performance of the proposed approach is shown on both artificial and real datasets.