Sensitivity Neural Network 1.0: Models for simulating the behaviour approach of artificial neural networks
Easy-to-use computer program that simulates the behaviour of artificial neural networks.
Artificial Neural Networks (ANN) are information processing systems with a structure and operation inspired by biological neural networks. They are characterized by being adaptive systems, that is, they learn to perform certain tasks through training with illustrative examples. As a result of this learning, the ANNs create their own internal representation of the problem and can respond not only to absolutely new information but also to distorted or incomplete information.
ANN can be used as a model for the study of the nervous system and cognitive phenomena or as a tool for predicting problems in different areas of knowledge (biology, medicine, etc.). In this second case, a methodology that presents common aspects with the conventional techniques of statistical modelling is applied. The simulation of the behaviour of the ANN has as an objective the explanation and prediction of phenomena from the study of the relationship between different variables.
The UIB offers:
- Training for the utilization of ANN as a statistical tool
- Training for handling the Sensitivity Neural Network 1.0 program
- Performing data analysis using ANN
The researchers of the UIB have successfully applied the ANN for the study of addictive behaviours. From surveys to 10,000 adolescents, a model of artificial neural networks has been constructed with the aim of finding characteristics that define drug users and evaluating the factors that prevent future drug addiction, to modify habits and risk circumstances