ABSTRACT

Recent advances in neuroscience, computer science, psychology, and other fields have led to the development of computer programs that are modeled, in principle, on idealizations of organic neural structures. These artificial neural networks (ANNs) exhibit important properties that promise great usefulness for policy researchers. Most important among these are ANNs'ability to learn to identify complex patterns of information and to associate them with other patterns. Furthermore, like their biological predecessors, ANNs can recognize and recall these patterns and associations in spite of noisy, incomplete, or otherwise defective information inputs. ANNs can also generalize information learned about one or more patterns to other related patterns. As a result, ANNs have already found extensive use in areas once reserved for multivariate statistical programs such as regression and multiple classification analysis, and are developing an extensive community of advocates for processing text and other qualitative information.