Feature Selection using Filter Methods - Tutorial 1

preview_player
Показать описание
During the Machine Learning Training pipeline we select the best features which we use to train the machine learning model.
In this video I explained the what is the Filter method and their different types. here I only explained the conceptual understanding about the filter method.
Below are different Filter Methods which I explained in a summarized way.

Filter Method Types

1. Basic Filter Methods
VarianceThreshod (Remove the Constant Feature and Quasi-Constant Features)
Remove Duplicate Features
2. Correlation & Ranking Filter Methods
Pearson’s correlation coefficient
Spearman’s rank coefficient
Kendall’s rank coefficient
3. Statistical Methods
Anova or F-Test
Mutual Information
Chi Square

#FeatureSelection #DataScience #MachineLearning
Рекомендации по теме
Комментарии
Автор

As South East Asian, It is usually difficult for me to follow tutorial from an Indian teacher, but you are not the case. Your explanation is very clear.

mrlajel
Автор

I like your way of explanation. It complete both Practical and theory..

rajbir_singh
Автор

Sir can u please how to calculate f-score feature selection method

sanheera