Evolved Analytics – Advanced, Proven, Pragmatic

Established in 2005, Evolved Analytics tackles real-world problem solving using nonlinear modeling. Founders, Dr. Mark Kotanchek and Dr. Theresa Kotanchek, spent decades in the science and engineering space, solving all sorts of cross-industry problems using a uniform workflow and unique genetic algorithms.

Evolved Analytics offers a complete portfolio of consulting, project design, execution and training that complements unique do-it-yourself software suites that continue to outperform other augmented analytics, artificial intelligence (AI) and machine learning platforms. Built to provide data-driven understanding of complex, unknown, nonlinear systems involving tens to thousands of input dimensions, Evolved Analytics’ solutions focus on the development, maintenance and deployment of transparent, robust, and interpretable input-response models. Our Analytics tools discover the most elusive relationships in input-response data.

Theresa Kotanchek

Theresa Kotanchek

Chief Executive Officer

Evolved Analytics, LLC
Read bio

Mark Kotanchek

Mark Kotanchek

Chief Technical Officer

Evolved Analytics, LLC
Read bio

What Our Users Are Saying

DATAMODELER PROVIDES QUICK TIME TO INSIGHT AND PUBLICATION-READY OUTPUT

“At Economy Monitor we evaluated genetic programming software from multiple vendors before choosing DataModeler. Evolved Analytics has created a unique technology that offers the flexibility of visualization-based exploration (suitable for data scientists) while allowing for rigorous statistical and scientific analysis. DataModeler simultaneously provides quick time to insight, and publication-ready output — thus, unrivaled research ROI.”

Percy Venegas, Chief Scientist, Economy Monitor

ONE OF THE MOST POWERFUL TOOLS FOR FINANCIAL DATA

“DataModeler is one of the most powerful tools available for analyzing financial data and constructing models which recognize hidden patterns amongst the avalanche of data that is used in trading. The reason for DataModeler’s advanced analysis powers rests with the high-level programming capability of Mathematica, on which it is built and for which it stands as an outstanding example of the type of complex tool that can be built using Mathematica.”

Dr. Michael Kelly, Wolfram Finance Technology Consultant, WRI, IL, U.S.A.

SUPPORTS THE DEMANDING WORKFLOWS OF SPECIALISTS AND SCIENTISTS

“DataModeler makes applied genetic programming accessible to practitioners across different disciplines, while being also robust to support the demanding workflows of specialists and scientists.”

Percy Venegas, Chief Scientist, Economy Monitor

FRIGHTENING SPEED AND EFFICIENCY

“DataModeler is one of those rare products that changes the way you think. It ends any excuse for extending an assumption of linearity in modeling beyond the domains in which it is truly appropriate. Using brilliant and flexible design that is itself the result of years of evolution, DataModeler harnesses natural selection to hunt down from the infinite computational universe and with frightening speed and efficiency ensembles of parsimonious non-linear models that predict well.”

Prof. Seth J. Chandler, Foundation Professor of Law, Director of the Program on Law and Computation, University of Houston Law Center, U.S.A.

INCREDIBLE SPEED AND OUT-OF-THE-BOX PERFORMANCE

“With DataModeler we were able to model a data set with 32 attributes and over 10,000 rows in less than an hour. The ensemble it produced was far more accurate than anything else we’ve seen. This is incredible out-of-the-box performance.”

Dr. Conor Ryan, Director at the Biocomputing and Developmental Systems Group in the Computer Science and Information Systems Department at the University of Limerick, Ireland

ROBUST CAPABILITIES FOR DIFFICULT PROBLEMS

“I have used it with success to explore optimal insurance contracts, issues in legal education and a host of other difficult problems. Often producing results that are superior to those from neural networks and that end up more amenable to traditional statistical analysis, DataModeler never fails to astound me with its robustness, its capabilities, and its ever-growing network of documentation.”

Prof. Seth J. ChandlerFoundation Professor of Law, Director of the Program on Law and Computation, University of Houston Law Center, U.S.A.

DATAMODELER IS A KING OF MACHINE LEARNING

“I started learning machine learning (ML) and deep learning. In a course, a sheet of data was given to do regression analysis. The data had so-called multiple co-linearity. I ran DataModeler for 3 hours and got a function plot [with] 1- R^2 equivalent with Scaled Vector Modeling. This value would be the lowest (best) one any ML can reach with the dataset. And it can be attained by anyone using DataModeler with little knowledge of ML. So, I’d like to call DataModeler a king of ML. It should be more widely used in the world.”

 Kazuhiro Iwadoh, Tokyo Women’s Medical University, Japan
Kazuhiro Iwadoh researches organ transplants

VARIABLE IMPORTANCE INSIGHTS FROM DATAMODELER

“We integrated consumer-oriented variable importance insights from DataModeler into a product design flow rapidly. DataModeler helped us identify the driving variable behind a consumer’s product preference in a matter of hours. From there, it was a rapid step to segment the consumers and design for each segment.”

Kalyan Veeramachaneni, Ph.D., Director, Data-to-AI, MIT, Cambridge, MA, USA

DATAMODELER IS AHEAD BECAUSE IT GENERATES TRUSTABLE MODELS

“The genetic programming approach for risk hedging will become central to algorithmic finance. First, idealized economic decision making models can not capture the complexity of human and human-machine behavior: only empirical time series models can. Second, in a world where information is abundant but attention and trustworthiness are scarce, value does not depend only on what we know, but on how much we can trust what we think we know. DataModeler is ahead because it generates trustable models.”

Percy Venegas, Chief Scientist, Economy Monitor

QUICKLY DETERMINE THE OPTIMAL MANUFACTURING PARAMETERS

“In a few hours, DataModeler gave us a set of explicit, non-linear models that improved our understanding of how key product variables changed with the manufacturing inputs. Other machine learning models which we have tried, are either linear and miss the non-linear interactions, or give a black box model that does not provide any learning.”

 Sweta Somasi, Manufacturing Technology Scientist, Corteva Agriscience