8000 mRMR (Minimum Redundancy and Maximum Relevance) score as score_func for feature selection methods · Issue #20067 · scikit-learn/scikit-learn · GitHub
[go: up one dir, main page]
Skip to content

mRMR (Minimum Redundancy and Maximum Relevance) score as score_func for feature selection methods #20067

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Closed
iki77 opened this issue May 9, 2021 · 1 comment
Closed

mRMR (Minimum Redundancy and Maximum Relevance) score as score_func for feature selection methods #20067

iki77 opened this issue May 9, 2021 · 1 comment
Labels
New Feature

Comments

@iki77
Copy link
iki77 commented May 9, 2021 •
edited
Loading

Describe the workflow you want to enable

Current score_func for feature selection methods does not consider multicollinearity between features.

Describe your proposed solution

Introduce mRMR (Minimum Redundancy and Maximum Relevance) score as score_func for feature selection methods.

Variant of mRMR scores in a nutshell:

  • MID: Mutual Information to target - Mutual Information between features
  • MIQ: Mutual Information to target / Mutual Information between features
  • FCD: F Statistic to target - Correlation between features
  • FCQ: F Statistic to target / Correlation between features

From what I understand Mutual Information and F Statistic already implemented as score_func in scikit-learn, so these mRmR scores are somewhat an extension of it.
@glemaitre
Copy link
Member

duplicate of #8889

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
New Feature
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@glemaitre @iki77
0