This year we are celebrating the 100th Anniversary of A.G. Ivakhnenko, the author of the unparalleled self-organizing, noise immune, inductive modeling and knowledge mining technology implemented in the Insights app, which has also been used for all the Global Warming forecasts presented here. He originated essential ideas found in many other data mining methods today, and having hosted Norbert Wiener, the father of cybernetics, after a major conference, Ivakhnenko authored more than 40 books and 500 scientific papers over his 65-year career, making significant contributions to informatics, cybernetics, artificial intelligence, intelligent control, and modeling.

He initiated the development of a proven noise immunity theory for inductive modeling not found in any other data mining technology today as a key feature of data analysis. It has been mathematically proven that models obtained by self-organizing, inductive modeling predict more accurately on noisy data than models based on physical principles and domain knowledge.

1. In his 1981 paper, Stanley J. Farlow effectively summarized the relevance of Ivakhnenko's GMDH, "A major difficulty in modeling complex systems in such unstructured areas as economics, ecology, sociology, and others is the problem of the researcher introducing his or her own prejudices into the model. Since the system in question may be extremely complex, the basic assumptions of the modeler may be vague guesses at best. It is not surprising that many of the results in these areas are vague, ambiguous, and extremely qualitative in nature.
   

2. 
   

3. "It was for this reason that in the mid 1960's the Ukrainian mathematician and cyberneticist, A.G. Ivakhnenko, introduced a method that allows the researcher to build models of complex systems without making assumptions about the internal workings. The idea is to have the computer construct a model of optimal complexity based only on data and not on any preconceived ideas of the researcher; that is, by knowing only simple input-output relationships of the system, Ivakhnenko's GMDH algorithm will construct a self-organizing model that can be used to solve prediction, identification, control synthesis, and other system problems."
   

Today, self-organizing inductive modeling is a proven and highly efficient knowledge extraction technology. Recent advances in research and development have made possible parallel implementations, multi-level self-organization for modeling high-dimensional data sets containing many thousands of input variables, cost-sensitive modeling, and new model evaluation techniques to improve the reliability and applicability of models. This technology has been employed successfully in various fields, from image recognition over biomarker detection and QSAR modeling, wastewater management and reuse questions, to Global Warming and micro and macro-economic forecasting problems.