Abstract
The ability to predict properties of molecules prior to their synthesis can be of great importance in optimizing their design. Substantial savings in time as well as cost can be achieved if some desired properties can be predicted prior to the synthesis of the molecule. The outer orbitals of a molecule are primarily responsible for many different properties of the molecules. These outer orbital parameters can be utilized by an intelligent computing system like an Artificial Neural Network to extract necessary knowledge about the properties of the molecule. An Artificial Neural Network was trained to extract electrical parameters of several polymers. The former was then tested using a number of molecules set aside (not used in training) solely for this purpose.
Recommended Citation
Mitra, Sanjay K.; Luo, Qing; and Darsey, Jerry A.
(1997)
"Artificial Neural Networks Used to Predict Electrical Properties of Polymers,"
Journal of the Arkansas Academy of Science: Vol. 51, Article 22.
Available at:
https://scholarworks.uark.edu/jaas/vol51/iss1/22
