Neural Networks and Their Applications

Abstract

The current interest in artificial neural networks can be attributed, in part, to the development of the modern computer. Since the advent of inexpensive, efficient, highstorage capacity computers, there has been an information explosion in many scientific disciplines as researchers are able to acquire larger and more comprehensive data sets. The interpretation of much of these data often requires manual inspection by scientists, especially when traditional methods of analysis do not appear to find the important relationships in the data. Manual inspection of data can be repetitive, time consuming, and difficult when many variables are involved simultaneously. Several novel processing schemes have been devised that attempt to supplement traditional signal processing techniques in difficult applications. One such approach for finding interesting relationships in multivariate data is the field of artificial neural networks (ANN).