Autonomous evolutionary algorithm in medical data analysis

An autonomous evolutionary algorithm for constructing decision trees is presented. The algorithm requires no or minimal human interaction and shows some interesting properties when used on different medical datasets. The algorithm uses a non-standard implicit fitness evaluation in the selection phase of a co-evolving environment. Together with self-adaptation of evolution parameters and with some other improvements it can monitor and adjust its own behavior. The algorithm's capability to self-adapt to a given problem is used as a measure to predict if some dataset is just difficult or impossible to analyze. The autonomous algorithm on average produces very general solutions or gives no solution if the dataset is prone to the overfitting problem.