Determining the significance of features with the use of Sobol method in probabilistic neural network classification tasks

In this article, the problem of determining the significance of data features is considered. For this purpose the algorithm is proposed, which with the use of Sobol method, provides the global sensitivity indices. On the basis of these indices, the aggregated sensitivity coefficients are determined which are used to indicate significant features. Using such an information, the process of features' removal is performed. The results are verified by the probabilistic neural network in the classification of medical data sets by computing model's quality. We show that it is possible to point the least significant features which can be removed from the input space achieving higher classification performance.