The Art of Feature Engineering
Users: 1 - Average Rating: 5.00
This book is a practical guide to feature engineering, which is an essential tool for machine-learning engineers to improve the performance of their models not by collecting more data or tuning the model’s parameters but modifying the data's features to better capture the nature of the problem.
The text starts from the basic concepts and techniques of feature engineering, then it studies data coming from different domains, graphs, texts, time series, and images, with relevant case studies and examples. Key topics of the text include binning, out-of-fold estimation, feature selection, encoding variable-length data, and dimensionality reduction.
The text starts from the basic concepts and techniques of feature engineering, then it studies data coming from different domains, graphs, texts, time series, and images, with relevant case studies and examples. Key topics of the text include binning, out-of-fold estimation, feature selection, encoding variable-length data, and dimensionality reduction.
Type:
Book
Area:
Data Analytics, Machine Learning
Target Group:
Basic
DOI:
https://doi.org/10.1017/9781108671682
Cite as:
Duboue, P., The art of feature engineering: Essentials for machine learning, Cambridge University Press, 2020.
Author of the review:
Giulia Cademartori
University of Genoa
You have to login to leave a comment. If you are not registered click here
Vincenzo DAmato DAmato
This is a practical guide to a topic that is critical to the effective use of Machine Learning algorithms: Featuring Engineering. The book, besides explaining the basics concept and main techniques of feature engineering, also includes case studies and coding examples.