Practical Fairness: Achieving Fair and Secure Data Models

This book provides practical guidance for data and AI professionals to develop fair and secure code by covering basic concerns related to data security and privacy, highlighting up-to-date academic research, and discussing legal developments.

Format: Paperback / softback
Length: 175 pages
Publication date: 18 December 2020
Publisher: O'Reilly Media, Inc, USA

The growing importance of fairness in data science is becoming increasingly evident. Mounting evidence suggests that the widespread adoption of machine learning and AI in business and government is perpetuating the same biases we aim to combat in the real world. However, when it comes to code, what exactly does fairness mean? This practical book addresses fundamental concerns related to data security and privacy, empowering data and AI professionals to utilize code that is fair, unbiased, and free from discrimination.

Numerous realistic best practices are emerging at every stage of the data pipeline, from data selection and preprocessing to closed model audits. Author Aileen Nielsen leads you through the technical, legal, and ethical aspects of creating code that is fair and secure, while highlighting up-to-date academic research and ongoing legal developments related to fairness and algorithms.

Identify potential bias and discrimination in data science models

Implement preventive measures to minimize bias during the development of data modeling pipelines

Understand which data pipeline components implicate security and privacy concerns

Write data processing and modeling code that follows best practices for fairness

Recognize the complex interrelationships between fairness, privacy, and data security created by the use of machine learning models

Apply normative and legal concepts relevant to evaluating the fairness of machine learning models

Weight: 614g
Dimension: 178 x 233 x 23 (mm)
ISBN-13: 9781492075738