Machine Learning by Tom M. Mitchell

Machine Learning Reviews

• 
  Ivan Idris

  This is an introductory book on Machine Learning. There is quite a lot of mathematics and statistics in the book, which I like. A large number of methods and algorithms are introduced:
  
  Neural Networks
  Bayesian Learning
  Genetic Algorithms
  Reinforcement Learning
  
  The material covered is very interesting and clearly explained. I find the presentation, however, a bit lacking. I think it has to do with the chosen fonts and lack of highlighting of important terms. Maybe it would have been better to have shorter paragraphs too.
  
  If you are looking for an introductory book on machine learning right now, I would not recommend this book, because in recent years better books have been written on the subject. These are better obviously, because they include coverage of more modern techniques. I give this book 3 out of 5 stars.
• 
  Anthony Singhavong

  Great intro to ML! For someone who doesn't have a formal Comp Sci background, this took a lot out of me. I found it helpful to stop after every chapter and listen to a more recent lecture to tie loose ends. Highly recommend reading this book in conjunction with professor Ng's ML intro course.
• 
  David

  This is a very compact, densely written volume. It covers all the basics of machine learning: perceptrons, support vector machines, neural networks, decision trees, Bayesian learning, etc. Algorithms are explained, but from a very high level, so this isn't a good reference if you're looking for tutorials or implementation details. However, it's quite handy to have on your shelf for a quick reference.
• 
  Samars

  This book is absolutely amazing. I love so much is my favorite book.
• 
  Lukas Simons

  🙄
• 
  Steve

  Great theoretically grounded intro to many ML topics.
• 
  Jacob Williams

  This is part of the required reading for the machine learning class in Georgia Tech's online master's program. I think it's weird that they use a book from the 90s, and that's especially annoying because it's expensive and relatively hard to find. It is a really well-written textbook, though.
• 
  Abhijit Gupta

  Read it back in 2018. I think it was my first book on ML. if you're well versed in probability, stats & linear algebra and want to get an outline of traditional ML, then this should serve as a nice weekend course on ML. :D
• 
  Terran M

  This book is a classic, but I can't stand it - to me it embodies everything wrong with how machine learning is often taught. ML people like to present the world from the point of view of optimizing a cost function for future examples, and see everything through this lens. This worldview can be useful for graduate-level research but it does not work for introductory teaching - it does not result in the student developing useful intuition, and people who learn in this way are unemployable.
• 
  Lurker

  Probably the first book you want in academic setting when studying machine learning. it's simple yet effective, and contains less mathematical mind-twisters and more concepts of machine learning algorithms.
• 
  Jethro Kuan

  A little dated, but had a nice way of introducing machine learning, classifying learning algorithms by their inductive biases. Would recommend other more modern books, such as the one by Kevin Murphy.
• 
  Niraj Upadhayaya

  A very nice introduction to Machine learning. It gives solid foundations. However very few examples are there.
• 
  Sushma

  Awesome book and so simply explained .
• 
  Eric Wallace

  Textbooks like this might not make for "fun" reading, but sometimes they're quite necessary. I was reading Tom Mitchell's classic machine learning survey book along with a machine learning survey class, as you might guess... and found it quite helpful. While the professor for my class made an effort to explain concepts and algorithms and such, I rarely began to understand the lectures until I read the textbook.
  
  One key point for reading this book: each chapter and section builds upon past content, so reading from the beginning is actually a good idea (despite what the intro says about jumping into what chapter you need). Mitchell is always careful to explain terminology and such so you don't necessarily need to look it up elsewhere, so if a term pops up that you're not familiar with, chances are good you missed the earlier explanation of it!
• 
  Brian Powell

  An easy, engaging text with a good selection of introductory topics from the field of machine learning. Mitchell covers decision trees, neural nets, Bayesian methods, rules and concept learning, and reinforcement learning, among others. Each is treated at just the right level: enough detail to chew on the concepts, but not a slog into the marginal particulars of this or that technique. Recommended as a good starter kit for those interested in machine learning: from here, you can launch off in a variety of directions, or supplement with other resources as needed. Only caveat is that it's a tad dated (and no coverage of unsupervised learning), though this is a minor nit for a lightweight introduction to the theory and concepts behind the various learning approaches.
• 
  Liuyang Li

  The book does a good job summarizing various research areas in machine learning. It, however, lacks enough formal treatments. It also does not provide enough details for each topics so that the read can fully understand the algorithms to apply them. I would recommend reading this book as an overview of machine learning and read books on each individual topic afterwards.
• 
  Todd Johnson

  Very clear prose. Covers an interesting sample of both probabilistic and non-probabilistic methods. Starting to feel a bit dated, as it does not cover important methods developed over the last decade, such as support vector machines. Nonetheless, the topics covered are covered very well.
• 
  Rasoul Nasiri

  A good introduction to machine learning, but I think it is not complete for learning machine learning.
• 
  DJ

  intro to machine learning
• 
  Hossein Kazemi

  Needs to be updated.