Non Linear Hebbian Learning techniques and Fuzzy Cognitive Maps in modeling the Parkinson's disease

A new soft computing method using Fuzzy Cognitive Maps for modeling and predicting Parkinson's disease has been proposed. A decision support system based on human knowledge and experience, with a Fuzzy Cognitive Map trained using unsupervised Nonlinear Hebbian Leanring algorithm are proposed. The basic theories of this learning method are reviewed and presented. The initial values of concepts are represented as fuzzy membership values and trained to get new updated weight matrix and new concept values. Simulations are performed and very interesting results are obtained and discussed. A comparison between the results with and without a learning algorithm is considered.