Artificial neural networks represent an excellent tool to develop real-world applications especially when traditional methods fail. Learning ability from data, classification capabilities, generalization and noise tolerance are few advantages that can be used successfully in the field of toxicity prediction. This paper focuses on the problem of describing the properties, effects or biological activities associated with chemicals through relations dependent on their structure, known as Quantitative Structure-Activity Relationship (QSAR) problem. We proposed a model using neural networks to predict the toxicity of chemicals with respect of three QSAR postulates: "the molecular structure is responsible for all the activities", "similar compounds have similar biological and chemo-physical properties" and "QSAR is applicable only to similar compounds".
CITATION STYLE
Crǎciun, M. V., Neagu, D. C., König, C., & Bumbaru, S. (2003). A study of aquatic toxicity using artificial neural networks. In Lecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science) (Vol. 2774 PART 2, pp. 911–918). Springer Verlag. https://doi.org/10.1007/978-3-540-45226-3_125
Mendeley helps you to discover research relevant for your work.