Genetic algorithm driven clustering for toxicity prediction
Dirk Devogelaere, Patrick Van Bael, M. J. Rijckaert · 2000
The pace of technological advancement in today’s society has generated an enormous demand for methods facilitating the intelligent testing for the toxicity of new chemicals. Until now it is common use to make prediction based on ‘real ’ tests. Recent investigations support the general assumption that macroscopic properties like toxicity and ecotoxicity strongly depend on microscopic features and the structure of the molecule. This paper’s authors have developed a computationally intelligent method for supervised training of regression systems, named GAdC (Devogelaere, 1999). Our method shall select those features needed to predict the toxicity and calculate the toxicity. The proposed methodology relies on supervised clustering with genetic algorithms and local learning.