Welcome to Bulletin of Botanical Research! Today is Share:

Bulletin of Botanical Research ›› 2014, Vol. 34 ›› Issue (6): 850-855.doi: 10.7525/j.issn.1673-5102.2014.06.023

Previous Articles     Next Articles

Discriminant Predictive Linear Regression Analysis Method for Astragalus

    

  1. 1.Modern Research Center for Traditional Chinese Medicine,Taiyuan 030006;
    2.College of Mathematics and Science,Shanxi University,Taiyuan 030006
  • Received:2014-03-11 Online:2014-11-20 Published:2015-01-06
  • Supported by:
     

Abstract: In this paper, the unary linear regression model between diameter of the xylem and cross section diameter was respectively set up for Astragalus membranaceus(Fisch.) Bge var. mongholicus(Bge.) Hsiao and A.membranaceus(Fisch) Bge.. Regression analysis and the classification of samples were conducted. Through several extracting random samples, correct classification rate was got. The results showed that the classification effect of A.membranaceus(Fisch.) Bge var. mongholicus(Bge.) Hsiao model was very good. In addition, the homogeneity of variance t test on samples of different growth years was carried on the samples of A.membranaceus(Fisch.) Bge var. mongholicus(Bge.) Hsiao. Through t test, it gave the confidence interval of the ratio between xylem and cross section diameter of samples of 2 years and above 2 years. Results showed that the interval of 2 years and above 2 years samples did not cross and could be obviously apart so that the growth years of Astragalus can be deduced. The study provides a simple and fast method for determining the varieties and growth years of Astragalus herbs. Hence, Astragalus samples can simply be classified and the growth years of a certain type of Astragalus can be determined.

Key words: Astragalus, classification, regression analysis, t test

CLC Number: