偏最小二乘法在食管鳞癌诊断中的应用

Application of Partial Least Squares Method in Diagnosis of Esophageal Squamous Cell Carcinoma (ESCC)

本文采用偏最小二乘法和主成分法分别对与食管鳞癌相关血清miRNAs数据进行降维处理,对降维数据利用传统的几种判别方法并结合十折交叉验证,得到了较高的分类准确率,对比两种方法的准确率,表明了在降维方面偏最小二乘法比主成法表现更好并分析了原因。

In this paper, partial least squares method and principal component method are used to reduce the dimension of serum microRNAs data associated with Esophageal Squamous Cell Carcinoma (ESCC), and several traditional discriminant methods and 10-fold cross-validation are employed to reduce the dimension of the data, and a higher classification accuracy is obtained. By comparing the accuracy of the two methods, it shows that the partial least squares method performs better than the principal component method in dimension reduction and the reasons are analyzed.