The modeling and optimization efficiencies of artificial neural network (ANN) and response surface methodology (RSM) in a two-step enzymatic hydrolysis of sweet potato was investigated in this study. Optimization of the process was carried out using RSM and generic algorithm (GA) of ANN which were then compared. The optimum reducing sugar yields predicted were 190.034 g/l and 244.6 g/l for liquefaction and saccharification, respectively. These compared well to ANN validated yield of 190.877 g/l and 244.68 g/l for liquefaction and saccharification, respectively. The ANN model R2 were 0.99998 and 0.99933 for both steps, respectively while 0.987 and 0.996 were obtained for both steps using RSM model. Also RMSE for ANN were found to be 0.1664 and 0.37922 while values for RSM were 3.19 and 0.58 for both steps. This showed that ANN had a higher predictive ability and was a better optimization tool than RSM on the hydrolysis of starch.
Cite This Article as:
[Adesina Olusola, Okewale Akindele, Olalekan Abiodun (2014); Comparative Studies of Response Surface Methodology (RSM) and Artificial Neural Network (ANN) Predictive Capabilities on Enzymatic Hydrolysis Optimization of Sweet Potato Starch Int. J. of Adv. Res. 2 (10). 0] (ISSN 2320-5407). www.journalijar.com
Share this article
This work is licensed under a Creative Commons Attribution 4.0 International License.