A Neuro-Classification Model for Socio-Technical Systems

This paper presents an original classifier model based on an artificial neural network (ANN) architecture that is able to learn a specific human behavior and can be used in different socio-economic systems. After a training process, the system can identify and classify a human subject using a list of parameters. The model can be further used to analyze and build a safe socio-technical system (STS). A new technique is applied to find an optimal architecture of the neural network. The system shows a good accuracy of the classifications even for a relatively small amount of training data. Starting from a previous result on adaptive forecasting, the model is enhanced by using the retraining technique for an enlarged data set.