作者: Wei Pan , Mei Wu , Zhenzhu Zheng , Longhua Guo , Zhenyu Lin
关键词: Sampling (statistics) 、 Partial least squares regression 、 Mathematics 、 Linear discriminant analysis 、 Artificial intelligence 、 Pattern recognition 、 Authentication 、 Pseudostellaria 、 Near-infrared spectroscopy
摘要: Pseudostellaria heterophylla is a very popular traditional Chinese medicine herb, also called "Taizishen." Discrimination of P. from different regions critical for ensuring the effectiveness drug use, because effects are diversity each other. To discriminate rapidly and effectively, model extracted by competitive adaptive reweighted sampling (CARS) was established. Original spectra in wave number range 10,000 to 4,000 cm-1 were acquired. Orthogonal partial least squares discriminant analysis (OPLS-DA) used establish suitable model. CARS performed extracting key variables. We found that near-infrared spectrum series samples analyzed Row-center-SG, CARS, OPLS-DA can effectively distinguish regions, accuracy satisfactory terms good discrimination rate. These results show be identify regions. PRACTICAL APPLICATION: According our research results, we draw conclusion may those places origin powder with similar appearance.