Machine Learning for Network and Cloud Engineers: Get ready for the next Era of Network Automation Paperback – 26 Jan. 2023

I bought this book because the author commented on one of my LinkedIn posts, so I looked at his profile saw his book and, after looking at the “Look inside” content, decided this looked like quite interesting (yes, I am a very boring person). I’d done a module on AI at MSc level which involved reading parts of books like Russell Norvig’s “AI – A Modern Approach”. I’d also read books like “AI for Autonomous Networks”, which was edited by Mazin Gilbert, because I am interested in how AI is applied in telecom networking. Antich’s book did not disappoint.The book is broken into fairly short, easily digestible chapters. It explains basic concepts like the distinction between AI and ML (the latter is a subset of the former) really clearly. It explains that in practice the challenge with ML is more about getting access to suitable data, not finding the right data science tools. It then goes on to provide actual use cases and explain how the author has implemented data science to solve them. In some cases, ML is not needed. It is a sledgehammer to crack a nut. Simple statistics (e.g., Z score) is good enough. In some cases, ML can produce poor results on the first iteration and a lot more experimentation with algorithms is needed to get something useful (AutoML helps here).I highly recommend this book to anyone looking at applying ML to network problems or indeed any problems as this is just the field Antich works in. His book will serve any reader looking to better understand ML and how it can be applied in practice to any field. If I have one criticism of the book it is that Antich did not infuse it with enough of his own humour. It helps to have a bit when dealing with fairly dry topics.