Abstract
There are more than 12,000 kinds of traditional Chinese medicine recorded in the dictionary, and different kinds of traditional Chinese medicine are distributed in many places of origin, so the task of identifying the place of origin of traditional Chinese medicine is extremely difficult. In this paper, the near-infrared spectral data and mid infrared spectral data provided by the 2021 “Higher Education Society Cup” National Undergraduate Mathematical Modeling Contest E question “Ap-pendix 3” are taken as research samples. First, the data provided are preprocessed, feature vectors are extracted, and dimensions are reduced. Then, support vector machine (SVM) algorithm is used to solve the problem and evaluation indicators are used to achieve 83.8% accuracy in training sets and 98.7% accuracy in test sets.