✳flâneur — a map of the web's best reading

MRMR - Minimum Redundancy Maximum Relevance — 1.8.3

feature-engine.trainindata.com · 2,679 words · saved by 1 readers

We use cookies to recognize your visits and preferences, to measure the effectiveness of the documentation, and discover how people find us. With your consent, you're helping us make our documentation better. Learn More. Section Navigation MRMR() selects features based on the Maximum Relevance Minimum Redundancy framework. In this framework, features with a strong relationship with the target (high relevance), but weak relationship with other predictor features (low redundancy) are favored and hence selected. The MRMR algorithm obtains a measure of relevance and a measure of redundancy, and then it assigns an importance score to each feature based on the difference or ratio between relevance and redundancy. After that, it selects the features with the highest scores. MRMR was first described in bioinformatics as a method to select features for microarray gene expression data, and then expanded and popularized by Uber in the context of marketing models. The MRMR feature selection algori

MRMR - Minimum Redundancy Maximum Relevance # MRMR() selects features based on the Maximum Relevance Minimum Redundancy framework. In this framework, features with a strong relationship with the target (high relevance), but weak relationship with other predictor features (low redundancy) are favored and hence selected. The MRMR algorithm obtains a measure of relevance and a measure of redundancy, and then it assigns an importance score to each feature based on the difference or ratio between relevance and redundancy. After that, it selects the features with the highest scores. MRMR was first d

Explore this link on the map →

saved by

related reading