A novel approach for olive leaf extraction through ultrasound technology : Response surface methodology versus artificial neural networks

Ilbay Z., Sahin S., Buyukkabasakal K.

KOREAN JOURNAL OF CHEMICAL ENGINEERING, vol.31, no.9, pp.1661-1667, 2014 (SCI-Expanded) identifier identifier


Response surface methodology (RSM) and artificial neural network (ANN) were used to evaluate the ultrasound-assisted extraction (UAE) of polyphenols from olive leaves. To investigate the effects of independent parameters on total phenolic content (TPC) in olive leaves, pH (3-11), extraction time (20-60 min), temperature (30-60 A degrees C) and solid/solvent ratio (500 mg/10-20 mL) were selected. RSM and ANN approaches were applied to determine the best possible combinations of these parameters. Box-Behnken design model was chosen for designing the experimental conditions through RSM. The second-order polynomial models gave a satisfactory description of the experimental data. Experimental parameters and responses were used to train the multilayer feed-forward networks with MATLAB. ANN proved to have higher prediction accuracy than that of RSM.