Neural Networks for the analysis of Common Mental Disorders Factors

The data used were generated by a cross-sectional study carried out in Olinda (Brazilian city) to investigate the specific links between social class and common mental disorders. The study was made on 621 individuals. All variables are ordinal and they have been codified for binary numbers. In this work, the data for training and testing the network were divided as follows: 50% of the patterns from each vintage were assigned to the training set, 25% were assigned to the validation set, and 25% were reserved to test the network, as suggested by Proben1. The patterns were normalized to the range [– 1.0,+1.0], for all network processing units implemented hyperbolic tangent activation functions. The variables use are age, literacy, years of schooling, own occupation, income distribution, housing conditions, possession of household appliances, marital status, working conditions.