What is feature engineering?

Introduction

Undoubtedly, big data are now rapidly expanding in all science and engineering domains. Resulting in a remarkable improvement in the performance of machine learning models. However, significant features are often collected or generated by different sensors and methods, causing a bad influence on the model accuracy and demanding high computational resources for handling the model.

Therefore, we need practical methods to build great and performant models even with these features. This course will talk about a popular and practical machine learning technique called feature engineering. We will go through pretty much everything you need to know about feature engineering in a hands-on approach with python.

Notably, we will learn how to manage dataset variables to extract meaningful information to catch as much insight as possible and prepare them to be used in machine learning algorithms.

What is feature engineering?

Feature engineering is the process of using raw data to create new features and variables that better represent the underlying problem to the predictive models, resulting in a performant machine learning model and improved prediction accuracy.

However, feature engineering can be iterative, time-consuming, and resource-intensive. It takes much effort to effectively process features in our datasets before actually using them in a model.

Feature engineering consists of many processes, to name a few:

• Filling missing values in a variable.
• Encoding categorical variables into numbers.
• Variable transformation.
• Creating or extracting new features from ones already available in your dataset

Throughout this course, you will learn how to use many of these methods to modify and transform your features.

Feature Engineering is part of a significant Machine learning pipeline, as shown in the illustration below: