I’ve read the book “Feature Engineering for Machine Learning”, by Zheng and Casari. It’s an interesting perspective on Machine Learning, where the books usually focus on the fancy part and forget about the real grind that is to get a ML product in production. The first chapter on itself, “The machine learning pipeline”, illustrates what other books about it doesn’t explain: the usual steps in making a machine learning product.
The book has a more friendly tone, with very easygoing explanations about tf-idf and PCA, for example. It also uses different ML algorithms for doing feature extraction, inclusing K-Means and neural networks for this task.
I’ll keep adding more as I go through it.