An electronic nose based on an array of 6 metal oxide semiconductor sensors was used, jointly with artificial neural network (ANN) method, to classify Pecorino cheeses according to their ripening time and manufacturing techniques. For this purpose different pre-treatments of electronic nose signals have been tested. In particular, four different features extraction algorithms were compared with a principal component analysis (PCA) using to reduce the dimensionality of data set (data consisted of 900 data points per sensor). All the ANN models (with different pre-treatment data) have different capability to predict the Pecorino cheeses categories. In particular, PCA show better results (classification performance: 100%; RMSE: 0.024) in comparison with other pre-treatment systems.
C. Cevoli, L. Cerretani, A. Gori, M.F. Caboni, T. Gallina Toschi, A. Fabbri (2011). Classification of Pecorino cheeses using electronic nose combined with artificial neural network and comparison with GC–MS analysis of volatile compounds. FOOD CHEMISTRY, 129, 1315-1319 [10.1016/j.foodchem.2011.05.126].
Classification of Pecorino cheeses using electronic nose combined with artificial neural network and comparison with GC–MS analysis of volatile compounds
CEVOLI, CHIARA;CERRETANI, LORENZO;GORI, ALESSANDRO;CABONI, MARIA;GALLINA TOSCHI, TULLIA;FABBRI, ANGELO
2011
Abstract
An electronic nose based on an array of 6 metal oxide semiconductor sensors was used, jointly with artificial neural network (ANN) method, to classify Pecorino cheeses according to their ripening time and manufacturing techniques. For this purpose different pre-treatments of electronic nose signals have been tested. In particular, four different features extraction algorithms were compared with a principal component analysis (PCA) using to reduce the dimensionality of data set (data consisted of 900 data points per sensor). All the ANN models (with different pre-treatment data) have different capability to predict the Pecorino cheeses categories. In particular, PCA show better results (classification performance: 100%; RMSE: 0.024) in comparison with other pre-treatment systems.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.