Automatic design of growing radial basis function neural networks based on neighboorhood concepts

Despite the reputation of RBFNs (Radial Basis Function Neural Networks), RBFN design is not straightforward since the efficiency of the model depends on many parameters. RBFNs often require many manual parameter adjustments, which is a serious weakness especially when they have to be used automatically. In this paper, a method to design RBFNs for classification problems is proposed, with a view to obtaining classification models rapidly by minimizing manual parameters, with performances very close to the best attainable from numerous trials. The RBFN can be initiated automatically via the use of advanced clustering algorithms adapted to supervised contexts to find preliminary cells. The final architecture is obtained via a growing process controlled by different mechanisms in order to find small and reliable RBF classifiers. A candidate pattern is selected for creating a new unit only if it produces a significant quadratic error while presenting a significant classification potential from its neighborhood properties. The efficiency of the method is demonstrated on artificial and real data sets from the field of chemometrics.