ABSTRACT
In machine learning, support vector machines (SVM) are supervised learning models associated with learning algorithms that analyze data used for classification and regression analysis. An SVM model is a representation of the examples as points in space, mapped so that the examples of the separate categories are divided by a clear gap that is as wide as possible. Two calibration algorithms, namely partial least squares regression (PLSR) and least-square support vector machine (LS-SVM) analyses, were used to correlate the extracted spectra of salmon samples with the reference tenderness values estimated by Warner–Bratzler shear force method to quantitatively predict tenderness of salmon fillets with a good performance. Spectral data were analyzed using PLSR and LS-SVMs to establish the calibration models. Principal component analysis was performed to obtain the principal component spectra of the foregoing derivative and raw spectra to reduce the reflectance spectral dimension.
