Home of real teaching & learning
  • Full support for teachers
  • Focus on critical thinking
  • Engaging classroom activities
  • Integrated student eBook
  • Assessed tasks / qBank
  • Practice exam questions

The InThinking Guarantee: Our sites are written by expert practitioners and not by AI

See our AI policy

Disclaimer: InThinking subject sites are neither endorsed by nor connected with the International Baccalaureate Organisation.

Don't miss out, find out!

A4.2.2 Feature selection

After a dataset has been cleaned, one practical question remains: which of its columns should the model actually see? A dataset can hold dozens or hundreds of features, and many carry no useful signal, repeat information already present, or actively mislead the model. Feature selection is the process of identifying and keeping only the most informative attributes, and it sits between data cleaning and model training....

Help