A New Data-Driven Deep Learning Model for Pattern Categorization using Fast Independent Component Analysis and Radial Basis Function Network. Taking Social Networks resources as a case

This paper investigates the categorization problem using Data Mining techniques. We present a new conceptual model, which is named FICARBFN, for classifying patterns by using Fast Fixed-Point Algorithm for Independent Component Analysis and Radial Basis Function Network. It uses an artificial neural network model to find a single consolidated categorization, which is composed of tree process, variables selection, categorization, and finally models selection. Our categorization model used a hybrid technique that combines the advantages of factorial analysis and Neural Network approaches. Comparative study and experimental results showed that our scheme optimized the bias-variance on the selected model and achieved an enhanced generalization for Social Networks patterns recognition.