The Art of Feature Engineering
Essentials for Machine Learning

'Pablo Duboue is a true grandmaster of the art and science of feature engineering. His foundational contributions to the creation of IBM Watson were a critical component of its success. Now readers can benefit from his expertise. His book provides deep insights into to how to develop, assess, combine, and enhance machine learning features. Of particular interest to advanced practitioners is his discussion of feature engineering and deep learning; there is a pervasive myth in the industry that deep learning and big data have made feature engineering obsolete, but the book explains why that is often incorrect for real-world computing applications and explains the relationship between building effective features and deep neural network architectures. The book engages with countless other basic and advanced topics in the area of machine learning and feature engineering, making it a valuable resource for machine learning practitioners of all levels of experience.' J. William Murdock, IBM

When working with a data set, machine learning engineers might train a model but find that the results are not as good as they need. To get better results, they can try to improve the model or collect more data, but there is another avenue: feature engineering. The feature engineering process can help improve results by modifying the data’s features to better capture the nature of the problem. This process is partly an art and partly a palette of tricks and recipes. This practical guide to feature engineering is an essential addition to any data scientist’s or machine learning engineer’s toolbox, providing new ideas on how to improve the performance of a machine learning solution.

Beginning with the basic concepts and techniques of feature engineering, the text builds up to a unique cross-domain approach that spans data on graphs, texts, time series and images, with fully worked-out case studies. Key topics include binning, out-of-fold estimation, feature selection, dimensionality reduction and encoding variable-length data. The full source code for the case studies is available on a companion website as Python Jupyter notebooks.

[Lire la suite]