New Version of Parabon Crush Software Enables Deep Analytics on Challenging Datasets

Parabon Computation, a provider of extreme-scale grid computing software and services, has announced the latest release of Parabon Crush, a massively scalable, statistical data mining and predictive modeling application that draws its power from Parabon's Frontier Grid Platform.  Now enhanced with a search algorithm developed for NASA, Crush uses the idle capacity of thousands of computers on a Frontier Grid to discover optimal statistical models within high-dimensional datasets that are too computationally challenging for traditional approaches. 

Crush statistical models can be used to identify hidden correlations, explain variability and predict future phenomena in practically any domain including genomic research to benefit healthcare, forecasting stock performance, and econometric and psychometric modeling

The latest version of Crush employs a novel evolutionary algorithm that searches arbitrarily large model spaces deeply and effectively without needing to exhaust them.  The so-called "opportunistic evolution (OE)" algorithm is a combination of genetic algorithms specifically designed to maximally leverage grid-scale capacity, enabling Crush to discover important statistical relationships that are practically impossible to find otherwise.

"The new algorithm uses a form of sampling strategy, but not random sampling. As the software samples candidate statistical models, it is also scoring them based on their ability to model known outcomes from actual data," Dr. Steven Armentrout, founder and CEO of Parabon Computation, tells 5 Minute Briefing. "High scoring models are kept for further use while low scoring ones are discarded. Then, as more new models are created, they start combining elements of the previously high scoring models, thereby evolving their accuracy over time."

The new release of Crush can be installed as a Microsoft Excel Add-in, enabling modeling runs entirely from within Excel. Crush can also be called from the command line, facilitating its incorporation into data mining workflows.  It supports a variety of regression models and has a pluggable architecture that makes it easy to add others.

For more information about Parabon Crush, go here.