论文部分内容阅读
中药指纹图谱数据具有变量数很大而样本数较小的特点,本文中采用拉格朗日求极值的方法导出一种新的适合用于处理这类数据的偏最小二乘算法。结果表明:所得到新的算法,在处理中药指纹图谱数据时,与传统的偏最小二乘算法比较,节省存储单元,计算量小,计算速度快,因而计算效率高。
Traditional Chinese medicine fingerprint data has the characteristics of large number of variables and small number of samples. In this paper, a new method of partial least-squares, which is suitable for processing such data, is derived by using the method of Lagrange for extremum. The results show that the new algorithm obtained in the processing of traditional Chinese medicine fingerprint data is compared with the traditional partial least squares algorithm to save the storage unit, the calculation is small, the calculation speed is fast, and the calculation efficiency is high.