主分量分析方法和二次判别分析方法应用于基因芯片数据分析

Applied the PCA dimensiona reduction and Quadratic Discriminant Analysis(QDA) on Gene Data Expression Analysis

本文主要采用主分量分析方法和二次判别分析(QDA)有监督分类的方法来对基因芯片(微阵列)数据进行分析。PCA是一种提取海量的数据有效特征的有效方法。可以获得与原来基因芯片数据更为接近的成分的提取特征的效果。实验表明采用PCA方法事先对数据处理不可以提高基因芯片数据分析的准确性。得出结论可为工业应用提供科学依据。

In the paper aims at the PCA o dimensiona reduction and Quadratic Discriminant Analysis（QDA） discriminant methods：to do the data analysis on gene chip（micoarray）. PCA and PLS, which have been developed recently, are efficient methods for analyzing numerous data. it can extract the features much closer to the gene data expression of origi- nality. It shows that when PCA is pre - disposal of the dataset, the accuracy of classification of gasoline is not improved naticeably . the writer makes an analysis conclusion and provides support for future industry.