15th European Conference on Artificial Intelligence
|
July 21-26 2002 Lyon France |
[full paper] |
Ciprian-Daniel Neagu, Emilio Benfenati, Giuseppina Gini, Paolo Mazzatorta, Alessandra Roncaglioni
In the present paper, models based on neuro-fuzzy structures are developed to represent knowledge about a large data set containing descriptors about organic compounds, commonly used in industrial processes. The neuro-fuzzy models here proposed include both, QSAR equations, and original numerical values. The proposed methods use various techniques to insert knowledge by training, and to map empiric fuzzy rules in neuro-fuzzy structures. The combinations of generalized fuzzy computation, neuro-fuzzy models, and strategies to insert data in the developed architectures, are used as a powerful processing tool in toxicity prediction.
Keywords: Knowledge-based Systems, Neural Networks, Knowledge Representation
Citation: Ciprian-Daniel Neagu, Emilio Benfenati, Giuseppina Gini, Paolo Mazzatorta, Alessandra Roncaglioni: Neuro-Fuzzy Knowledge Representation for Toxicity Prediction of Organic Compounds. In F. van Harmelen (ed.): ECAI2002, Proceedings of the 15th European Conference on Artificial Intelligence, IOS Press, Amsterdam, 2002, pp.498-502.