DataModeler in Epidemiological Studies

Finding relationships in real data and drawing reliable predictions from them is hard. Evolved Analytics’ DataModeler elegantly handles challenges of the real world - superfluous factors, correlated inputs, nonlinear couplings, noise & measurement uncertainty, missing data, redundant data, too many factors and not enough data.

Non-linear modeling with built-in automatic variable selection, robust ensemble-based prediction that comes with quantified uncertainty, transparent, human- interpretable models is what makes the cutting-edge technology behind DataModeler a perfect analysis tool for epidemiological studies.

Contrary to other data mining methods, Symbolic Regression models as implemented in DataModeler are explicit mathematical expressions which are designed to have minimal complexity and automatically focus on the driving factors only.

The comprehensive model analysis environment helps epidemiologists explicitly quantify the absolute and relative importance of factors in the data without making simplifying and often constraining assumptions. Experienced statisticians love DataModeler because it is such a great hypothesis generator.