From 3f044549ca49a70a93850fbea5964460a9f8e0f3 Mon Sep 17 00:00:00 2001 From: Steve Driscoll Date: Sun, 20 Mar 2022 18:06:35 -0300 Subject: [PATCH] Update README.md --- README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README.md b/README.md index 6e55689..fd8efaf 100644 --- a/README.md +++ b/README.md @@ -4,7 +4,7 @@ Kurtosis-based Projection Pursuit Kurtosis-based projection pursuit analysis (PPA) is an exploratory data analysis algorithm originally developed by Siyuan Hou and Peter Wentzell in 2011 and remains an active research area for the [Wentzell Research Group](http://groupwentzell.chemistry.dal.ca/) at Dalhousie University. Instead of using variance and distance-based metrics to explore high-dimensional data (PCA, HCA etc.), PPA searches for interesting projections by optimizing kurtosis. This repository contains MATLAB and Python code to perform PPA, published literature involving the on-going development of PPA, as well as some examples of how to apply PPA to uncover interesting projections in high-dimensional data. Below is a figure from our recent paper that I think demonstrates the value of searching for distributions with low kurtosis values.

-kurtosis +kurtosis