Potential function based network

D. Zhao, W. Chen, and S. Hu, Comput. Chem., 22, 385 (1998). Potential Function Based Neural Networks and Its Applications to the Classification of Complex Chemical Patterns. 

Having discussed self-assembly strategies toward noncovalently functionalized side chain supramolecular polymers as well as studies toward the orthogonahty of using multiple noncovalent interactions in the same system, this section presents some of the potential applications of these systems as reported in the literature. The apphcations based on these systems can be broadly classified into two areas 1) self-assembled functional materials and 2) functionalized reversible network formation. 

KNN)13 14 and potential function methods (PFMs).15,16 Modeling methods establish volumes in the pattern space with different bounds for each class. The bounds can be based on correlation coefficients, distances (e.g. the Euclidian distance in the Pattern Recognition by Independent Multicategory Analysis methods [PRIMA]17 or the Mahalanobis distance in the Unequal [UNEQ] method18), the residual variance19,20 or supervised artificial neural networks (e.g. in the Multi-layer Perception21). 

The variance approach may also be used to determine n. From Illustration 11.2 the variance of the response data based on dimensionless time is 30609/(374.4)2, or 0.218. From equation 11.1.76 it is evident that n is 4.59. Thus the results of the two approaches are consistent. However, a comparison of the F(t) curves for n = 4 and n = 5 with the experimental data indicates that these approaches do not provide very good representations of the data. For the reactor network in question it is difficult to model the residence time distribution function in terms of a single parameter. This is one of the potential difficulties inherent in using such simple models of reactor behavior. For more advanced methods of modeling residence time effects, consult the review article by Levenspiel and Bischoff (3) and textbooks written by these authors (2, 4). 

CNTs own excellent materials properties. DNA is an excellent molecule to construct macromolecular networks because it is easy to synthesize, with a high specificity of interaction, and is conformationally flexible. The complementary base-paring properties of DNA molecules have been used to make two-dimensional crystals and prototypes of DNA computers and electronic circuits (Yan et al., 2002 Batalia et al., 2002). Therefore functionalization of CNTs with DNA molecules has great potential for applications such as developing nanodevices or nanosystems, biosensors, electronic sequencing, and gene transporters. 

Molecular mechanics is a simple technique for scanning the potential energy surface of a molecule, molecular ion, crystal lattice or solvate. The model is based on a set of functions which may or may not be based on chemical and physical principles. These functions are parameterized based on experimental data. That is, the potential energy surface is not computed by fundamental theoretical expressions but by using functions whose parameters are derived empirically by reproducing experimentally observed data. Molecular mechanics then is, similar to a neural network, completely dependent on the facts that it has been taught. The quality of results to be obtained depends on the choice of the experimental data used for the parameterization. Clearly, the choice of potential energy functions is also of some importance. The most common model used is loosely derived from... 

---