The relationship between the seismic data and the reservoir properties can be modeled by using statistical approaches, such as regression and artificial neural networks (ANN); however, another nonlinear regression method, known as the group method of data handling (GMDH), has been proven to perform better than regular statistical methods. GMDH is a supervised machine learning tool that automatically self-organizes (synthesizes) the models. Although it is self-organized, like unsupervised ANN, it learns from the examples introduced similar to the supervised ANN. We apply the GMDH algorithm to seismic attributes to predict reservoir porosity. GMDH can automatically determine the best network structure, as well as the number of nodes, thus gauging sensitivity of the input without overtraining the data. Moreover, GMDH predicted porosity has better resolution than that predicted using ANN.