Automated derivation of reaction rules for the EROS 6.0 system for reaction prediction

The purpose of the EROS 6.0 system is to predict the products of chemical reactions and to model reaction mechanisms. This is accomplished by elementary reaction steps that are selected through the rules that constitute the knowledge base of the EROS 6.0 system. These rules are derived by methods of machine learning. The learning process is based on reaction in data bases. An overview of the EROS 6.0 system is given and the structure of the knowledge as well as the generation of reaction rules are described.