Objective: To establish a near infrared spectroscopy method for rapid detection the characteristic contents in Wujiapi liquor extract. Method: The contents of glycyrrhizic acid

costunolide and dehydrocostuslactone determined by HPLC method were used as reference values and the near infrared spectroscopy of the Wujiapi liquor extract were collected by fourier transform near infrared spectroscopy technology. Partial least square (PLS) was utilized to build the quantitative calibration models and the unknown samples were quantitatively determined. Result: The results showed that the mean square errors of cross-validation (RMSECV) were 0.023 9

0.021 8 and 0.012 6

respectively

the correlation coefficient (R2) were 0.978 0

0.978 7 and 0.975 7

respectively. External validation samples collected were used to evaluate the quantitative models. The root mean square errors of prediction (RMSEP) were 0.023 7

0.018 8 and 0.010 7

respectively. Conclusion: Using near infrared spectroscopy the models have good predictive effect

which could meet the demand to rapid detect the characteristic constituents in Wujiapi liquor extract.