ABSTRACT
Mathematical literature reveals that the number of neural network structures, concepts, methods, and their applications have been well known in neural modeling literature for sometime. There is rapid development in artificial neural network modeling, mainly in the direction of connectionism among the neural units in network structures and in adaptations of “learning” mechanisms. This chapter presents differences and commonalities among inductive-based learning algorithms, deductive-based adaline, and backpropagation techniques. The focus here is on the presentation of emperical analyzing capabilities of the networks; i.e., multilayered inductive technique, adaline, backpropagation, and self-organization boolean logic technique, to represent the input-output behavior of a system. The aspects considered are: basic functioning at unit-level based on these approaches connectivity of units for recognition and prediction type of problems.
