Skip to main content

Introduction

Introduction
Complete

Data into Insights: The Power of Features​

Imagine a dataset with a thousand columns and one million rows, how would you feel about using it to train a machine learning model? You might wonder if all of it will be useful, and if not, how will you decide?

An effective model doesn’t happen just from having more data. It comes from the right data. This is what feature engineering helps us do.

Hand pointing illustration

Why does this unit matter?​

When we build machine learning models for our organisations, we need them to be as effective and efficient as possible.

Selecting the right features ensures the data is in its best format for our model. This leads to a system that will produce insights faster, allowing our organisations to make better decisions that promote growth.

Learning objectives​

By the end of this unit, you will be able to:

  • Define the importance of feature engineering, selection and preprocessing.
  • Analyse variable types (continuous, categorical, ordinal) and their impact on performance.
  • Implement selection methods that optimise performance and efficiency.
  • Apply transformation techniques to clean and prepare raw data.

Action item: Pause and think​

Before diving into the unit, take a moment to reflect on your own experience working with data. Feature engineering is where raw data becomes meaningful.

Questions & Reflections
1. Have you ever worked with a dataset that felt overwhelming or underwhelming? What did you do to make it usable?

Type your reflection here...

2. Think about a time when you made a spreadsheet or database easier to analyse. How might that relate to what ML models need?

Type your reflection here...