4. Feature selection using Correlation Threshold #machinelearning #deeplearning #datascience

preview_player
Показать описание
Welcome to our video on feature selection using correlation threshold! In this tutorial, we'll be exploring a technique for identifying and removing highly correlated features from a dataset. This method can be particularly useful when working with datasets that contain many redundant or highly correlated features.

We'll be demonstrating how to apply this technique in Python using pandas and scikit-learn. We'll start by loading and preparing the dataset, then we'll use pandas' .corr() method to calculate the pairwise correlations between features. Next, we'll use scikit-learn's SelectKBest method to select the top performing features based on their correlation score. Finally, we'll evaluate the impact of feature selection on the model's performance.

By the end of this video, you'll have a solid understanding of how to use correlation threshold to select relevant features and improve the performance of your machine learning models.

#featureselection #correlationthreshold #datapreprocessing #machinelearning #python #pandas #scikitlearn #selectkbest #modelperformance #correlatedfeatures #datacleansing #datawrangling
Рекомендации по теме
Комментарии
Автор

hey this was clear and nicely presented! thank you!

genie
Автор

what if the corr between 2 feature (X) are negative, should we drop each other or keep them both?

mujirah