Skip to Main content Skip to Navigation
Journal articles

Artificial Neural Network-Based Equation to Predict the Toxicity of Herbicides on Rats

Abstract : The use of herbicides is increasing around the world. The benefits achieved by the use of these herbicides are indisputable. Despite their importance in agriculture, herbicides can be dangerous to the environment and the human health, depending on their toxicity, and the degree of contamination. Also, it is essential and evident that the risk assessment of herbicides is an important task in the environmental protection. The objective of this work was to investigate and implement an Artificial Neural Network (ANN) model for the prediction of acute oral toxicity of 77 herbicides to rats. Internal and external validations of the model showed high Q2 and r m 2 - values, in the range 0.782 – 0.997 for the training and the test. In addition, the major contribution of the current work was to develop artificial neural network-based equation to predict the toxicity of 13 other herbicides; the mathematical equation using the weights of the network gave very significant results, leading to an R2 value of 0.959. The agreement between calculated and experimental values of acute toxicity confirmed the ability of ANN-based equation to predict the toxicity for herbicides that have not been tested as well as new herbicides
Document type :
Journal articles
Complete list of metadatas

Cited literature [69 references]  Display  Hide  Download

https://hal-univ-rennes1.archives-ouvertes.fr/hal-01295518
Contributor : Laurent Jonchère <>
Submitted on : Friday, June 10, 2016 - 12:49:19 PM
Last modification on : Thursday, March 5, 2020 - 2:02:51 PM

File

Artificial Neural Network-Base...
Files produced by the author(s)

Identifiers

Citation

Mabrouk Hamadache, Salah Hanini, Othmane Benkortbi, Abdeltif Amrane, Latifa Khaouane, et al.. Artificial Neural Network-Based Equation to Predict the Toxicity of Herbicides on Rats. Chemometrics and Intelligent Laboratory Systems, Elsevier, 2016, 154, pp.7-15. ⟨10.1016/j.chemolab.2016.03.007⟩. ⟨hal-01295518⟩

Share

Metrics

Record views

334

Files downloads

215