The Screening of Outliers in Data Mining with the Help of the χ~2 Test and tne Grubbs Test

摘要

With the help of the χ~2 test and the Grubbs test, this paper presents a synthetic method to screen data. First, we use the χ~2 test to judge whether the data set is normally distributed. If not, we can then use the Grubbs test to reject outliers step by step, so as to ensure the accuracy of the experimental data set.