Neural network system for knowledge discovery in distributed heterogeneous data

Abstract

This paper proposed a distributed KDD system allowing remote usage of expert knowledge. Its implementation on the base of polynomial neural networks is described. The system is an universal KDD tool as it can build decision-making models in any subject field. Its implementation as a webservice will allow third-party software developers to create specialized applications, which focus on neural knowledge base usage. © 2006 International Federation for Information Processing.