Feature Preparation, Selection and Engineering

Improve your Kaggle score by selecting the best features and creating new ones.


  • Learn how to determine which features in your model are the most-relevant to your predictions.
  • Learn ways to reduce the number of features used to train your model and avoid overfitting.
  • Learn techniques to create new features to improve the accuracy of your model.

Mission Outline

1. Introduction
2. Preparing More Features
3. Determining the Most Relevant Features
4. Training a model using relevant features.
5. Submitting our Improved Model to Kaggle
6. Engineering a New Feature Using Binning
7. Engineering Features From Text Columns
8. Finding Correlated Features
9. Final Feature Selection using RFECV
10. Training A Model Using our Optimized Columns
11. Submitting our Model to Kaggle
12. Next Steps
13. Takeaways


Course Info:


The median completion time for this course is 5.9 hours. View Details

This course requires a premium subscription, and includes three missions and one guided project.  It is the 28th course in the Data Scientist in Python path.


Take a Look Inside