Artificial Neural Networks and Their Applications

Abstract

Artificial Neural Networks (ANNs) are computational models inspired by the central nervous system, and they are capable of machine learning and pattern recognition. In addition, ANNs are increasing in popularity for solving many industry, business, and science problems. For example, ANNs are used in automotive systems, manufacturing quality control, financial prediction, and in the medical field—spotting cancerous cells. ANN models are composed of many nonlinear computational elements operating in parallel, and usually are presented as systems of interconnected neurons that can compute values from inputs by feeding information through the network. This work presents the following six popular ANN models: HopField Net, Hamming Net, Carpenter/Grossberg classifier, Single layer Perceptron, Multi-layer Perceptron and Kohonen Self-Organizing Feature. These ANNs are highly parallel building blocks that illustrate neural-net components, design principles, and can be used to construct more complex systems. Classification and clustering are major tasks in Knowledge Discovery and Data Mining. This work also discusses how some classification and clustering algorithms can be performed using these simple neuron-like components, and the feasibility of ANNs in many other application fields.