Flow of information through artificial neural network of chlorophenol QSAR model

The analysis of information flow by an artificial neural network (ANN) were developed based on the model study of Quantitative Structure-Activity Relationship (QSAR) of chlorophenol. By comparing the values of connection weights and biases of the ANN model, the key nodes of hidden and input layers for the nodes of output layer were selected, and the distribution figure of information flow through the ANN model was shown. It is useful to discover the more information from ANN model, and to study further on the quantitative structure-activity relationship and toxicological mechanism of organic chemicals.