Comparison of Multiple linear Regression and Neural Network for the Prediction of Entropy of Amino Acid Derivatives

The relationship between topological indices such as Randic, Balaban to the entropy (S) of 96 amino acid derivatives is represented. The quantum chemical parameter is taken from the quantum mechanics methodology with HF level using the ab initio 6-31G basis sets. The multiple linear regressions (MLR) and artificial neural network (ANN) were tested to give the QSPR models. Comparison of the results revealed that the application the ANN method gave better results than MLR method.