Integrating rules and neural nets for carcinogenicity prediction
Giuseppina C. Gini, Marco Lorenzini, Emilio Benfenati, Raffaella Brambilla, Luca Malvé · 2002
One approach to deal with real complex systems is to use more techniques in order to combine their different strengths and overcome each other's weakness to generate hybrid solutions. In this project we pointed out the needs of an improved system in toxicology prediction. An architecture able to satisfy these needs has been developed. The main tools we integrated are rules and ANN. We defined chemical structures of fragments responsible for carcinogenicity according to human experts, developing a module able to recognize these fragments in a chemical. Furthermore, we developed an ANN, using molecular descriptors as inputs to predict carcinogenicity as a numerical value. Finally, we developed an automatic learning program to combine the results into a classifications of carcinogenicity to man.