Classification for Glucose and Lactose Terahertz spectra based on SVM and DNN methods

We propose an approach based on support vector machine(SVM) and deep neural networks (DNN) to classify chemical substances under different experimental conditions in terahertz time-domain spectroscopy (THz-TDS). 372 groups of independent signals under different conditions were measured to provide a sufficient training set. 99% accuracy for the SVM and 89.6% for the DNN method are realized in the test set. These excellent classification results show the high potentials in chemical recognition, security detection or clinical diagnosis.