Neural and neuro-fuzzy models of toxic action of phenols

Author

Neagu, Ciprian-Daniel N. ; Aptula, Aynur O. ; Gini, Giuseppina

Author_Institution

Dept. of Comput. Sci. & Eng., Univ. Dunarea de Jos of Galati, Romania

Volume

1

fYear

2002

fDate

2002

Firstpage

283

Abstract

The problem of describing the bio-chemical action of different classes of chemical compounds through relations dependent on their structures is known as the quantitative structure-activity relation (QSAR) problem. Development of toxicity models of phenols using neural and neuro-fuzzy models is here proposed. A dataset about the inhibition of growth determined by phenolic compounds to the protozoan ciliate Tetrahymena pyriformis was used to produce QSAR and connectionist models. The results are promising, and suitable for further research.

Keywords

biochemistry; biology computing; chemistry computing; fuzzy neural nets; inference mechanisms; Tetralrymena pyrifortnis; biochemical action; chemical compounds; connectionist models; neural models; neuro-fuzzy models; phenotic compounds; protozoan ciliate; quantitative structure-activity relation problem; toxicity models; Artificial intelligence; Chemical analysis; Fellows; Fuzzy neural networks; Medical diagnosis; Natural languages; Neural networks; Predictive models; Speech; Toxic chemicals;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent Systems, 2002. Proceedings. 2002 First International IEEE Symposium

Print_ISBN

0-7803-7134-8

Type

conf

DOI

10.1109/IS.2002.1044269

Filename

1044269