Search icon
Subscription
0
Cart icon
Close icon
You have no products in your basket yet
Arrow left icon
All Products
Best Sellers
New Releases
Books
Videos
Audiobooks
Learning Hub
Newsletters
Free Learning
Arrow right icon
Data-Centric Machine Learning with Python

You're reading from  Data-Centric Machine Learning with Python

Product type Book
Published in Feb 2024
Publisher Packt
ISBN-13 9781804618127
Pages 378 pages
Edition 1st Edition
Languages
Authors (3):
Jonas Christensen Jonas Christensen
Profile icon Jonas Christensen
Nakul Bajaj Nakul Bajaj
Profile icon Nakul Bajaj
Manmohan Gosada Manmohan Gosada
Profile icon Manmohan Gosada
View More author details

Table of Contents (17) Chapters

Preface 1. Part 1: What Data-Centric Machine Learning Is and Why We Need It
2. Chapter 1: Exploring Data-Centric Machine Learning 3. Chapter 2: From Model-Centric to Data-Centric – ML’s Evolution 4. Part 2: The Building Blocks of Data-Centric ML
5. Chapter 3: Principles of Data-Centric ML 6. Chapter 4: Data Labeling Is a Collaborative Process 7. Part 3: Technical Approaches to Better Data
8. Chapter 5: Techniques for Data Cleaning 9. Chapter 6: Techniques for Programmatic Labeling in Machine Learning 10. Chapter 7: Using Synthetic Data in Data-Centric Machine Learning 11. Chapter 8: Techniques for Identifying and Removing Bias 12. Chapter 9: Dealing with Edge Cases and Rare Events in Machine Learning 13. Part 4: Getting Started with Data-Centric ML
14. Chapter 10: Kick-Starting Your Journey in Data-Centric Machine Learning 15. Index 16. Other Books You May Enjoy

Types of bias

In machine learning, there are generally five categories of bias that warrant attention. Although the list provided isn't exhaustive, these categories represent the most prevalent types of bias, each of which can be further subdivided.

Easy to identify bias

Some types of bias can be easy to identify using active monitoring and by conducting analysis. These include the following.

Reporting bias

This type of bias occurs when the data producers, data annotators, or data capturers miss out on important elements, which results in data not being representative of the real world. For instance, a healthcare business might be interested in patients’ sentiments toward a health program; however, the data annotators may decide to focus on negative and positive sentiments, and sentiments that were neutral may be underrepresented. A model trained on such data will be good at identifying positive and negative sentiments but may fail to accurately predict neutral...

lock icon The rest of the chapter is locked
Register for a free Packt account to unlock a world of extra content!
A free Packt account unlocks extra newsletters, articles, discounted offers, and much more. Start advancing your knowledge today.
Unlock this book and the full library FREE for 7 days
Get unlimited access to 7000+ expert-authored eBooks and videos courses covering every tech area you can think of
Renews at $15.99/month. Cancel anytime}