Non-math-modelling data analytical methods base on artificial neural networks(ANNs) applied to optimize preparation of norcantharidin-loaded chitosan nanoparticles

The artificial neural networks (ANNs) non-modeling methods were selected to optimize the preparation of loading norcantharidin chitosan nanoparticles (NPs) by ionic cross-linkage. A multiple regression model was constructed for fitting several preparation factors and each of the factor level values was arranged in the L9(34) design table and their linear weighted sum of the normalized value was taken as optimized object. A Back-Propagation (BP) network (3x7x2) in ANNs was created and trained for further checking the optimal results and the trained network was applied to simulate the experiment system and screen the optimal conditions. Finally, the best condition was obtained.