Аннотация:One of the important directions of research in geophysical electrical prospecting is solution of inverse problems (IP), in particular, the IP of magnetotellurics – the problem of determining the distribution of electrical conductivity in the thickness of earth by the values of electromagnetic field induced by ionosphere sources, observed on earth surface. One of the ways to overcome instability of this IP is simplification of the model of the section with its subsequent resolution. In this case, it is possible to limit oneself with solving a small number of known to be more stable IP in narrower classes of geoelectrical sections. This report presents the results of use of neural network based classification to determine the one of these narrow classes to which the result of the next observation belongs.