ABSTRACT
This chapter proposes a novel negentropy sorted kernel independent component analysis (k-ICA) method as a feature extraction method for Fourier-transform infrared (FTIR) spectroscopy. The kernel support vector machine (k-SVM) model has the best performance, and it is better than back propagation-artificial neural network and partial least squares. The improved double kernel method (k-ICA and k-SVM) can detect food’s variety, brand, origin and adulteration simultaneously, and the recognition performance is steady, high, and efficient, and the recognition program works steadily efficiently. Support vector machine (SVM) classification is a non-parametric supervised classification technique. Often SVM-based classifiers are shown to perform well in small sample research. The salient feature of SVM is the ability to handle data with very high dimensions with very little training samples. SVM has been widely used in many fields and can solve both linear and nonlinear multivariate calibration problems.
