Features selection using Kendall’s Rank Correlation

Question

[yoshifumimiya]

Hi, everyone!

I have used ANOVA for feature selection. →　Feature Selection using ANOVA

However, the variables (transcripts) were non parametrical data. In this case, I thought it would be better to try feature selection using Kendall's Rank Correlation.

However, I don't know what degree of Kendall's tau to use as a basis for feature selection.

I would appreciate your opinion.