Use of cluster analysis and rough set approaches in predicting toxicity of chemicals

• Autorzy: Dębska, B.
• Języki publikacji: PL

Abstrakty

EN

The paper presented the applications of quantitive structure-activity relationships (QSAR) in environmental chemistry and toxicology. In this study, statistical and rough set methods have been applied to the development of QSAR models for estimating the acute aquatic toxicity of selected chemical compounds. Physicochemical and topostructural indices were used as properties relevant to the assessment of toxicity. These parameters have been used in the formulation of QSAR models for predicting toxicological properties. In the experimental part of the paper, results obtained in testing of ROSETTA toolkit and VVT module are presented. The rough set methodology were then used in selection and reduction the number of paramaters (chemical indexes described the selected compounds) before cluster analysis. Cluster analysis methods were used in unsupervised learning, classifying new compounds and predicting toxicity of chemicals. The results of test cases show that both give accaptable estimates for the toxicity of the compounds studied in this paper.

Słowa kluczowe

EN

knowledge dicsovery; QSAR; roughs set theory; cluster analysis methods

• Czasopismo: Prace Instytutu Podstaw Informatyki Polskiej Akademii Nauk
• Rocznik: 2001
• Tom: Nr 928
• Strony: 12-19

Twórcy

autor

Dębska, B.

• Department of Computer Chemistry Rzeszow University of Technology ul. Powstańców Warszawy 6 35-041 Rzeszów,