Comparison between Artificial Neural Networks and Support Vector Machine Modeling for Polycaprolactone Synthesis via Enzyme Catalyzed Polymerization

In this research, comparison between artificial neural network (ANN) and support vector machine (SVM) techniques for prediction of bio-polymer molecular weight is dealt with. Synthesis of polycaprolactone (PCL) via enzymatic catalysis has been chosen for this study. Initially, the process parameters for enzymatic synthesis of PCL are optimized by using D optimal method. Machine learning (ML) techniques have been applied to predict the output molecular weight of the biopolymer synthesized via enzymatic polymerization. Two popular ML algorithms, support vector machine (SVM) and artificial neural network (ANN), have been proposed to evaluate the polymer molecular weight as output. The performance of selected solutions leads to a closed prediction with the real experiments. Results from ANN and SVM techniques are also compared with various ML techniques and statistical forecasting methods. The results from both training and testing samples indicate SVM as a proper solution with respect to the characteristic of polymerization problem.