I have this book listed as one of the best and most interesting I've ever read. I loved the book just as much as I loved the course we used it in.

I have a genuine interest in AI and especially Machine Learning and this book both inspired me as well as clared some things up for me. Like how the spectrum of different known methods differ in their appoach of different problems.

This book is also very concise and well written, without being too mathematical. Making it very easy to read and understand.

Ever since I took that course I keep returning to this book as a reference when I have a related problem to solve, or just bothering my mind.

Highly recommended!

great book if you wanna start sth anywhere in machine learning, but it is toooooo expensive.

This is a great book if you're starting out with machine learning. It's rare to come across a book like this that is very well written and has technical depth. The writing is to the point, maybe even a bit terse, but all that you need to know is in there. It's a bit old so doesn't cover kernel methods or SVM's but is still a great first machine learning book.

Intelligent machines can be characterized, at least qualitatively, by the degree to which they require interaction with another intelligent machine in order to learn or engage in problem solving, the degree to which they can construct theories explaining data in a given domain, the degree to which they can think in more than one domain, the degree of curiosity that they possess, and the degree to which they take on these tasks by themselves and not under the instigation of an external entity. There are many (non-human) machines today that have some of these characteristics, thanks in part to the research contained in this book, among many others.

The book is a collection of work in artificial intelligence that was undertaken almost 25 years ago, but some of it is still relevant today. At that time researchers were still unsure as to what constitutes machine intelligence, and to a large degree the lack of a general definition has continued to this day. But one can argue with confidence that machine intelligence has finally taken a foothold in the real world, and in fact is now undergoing an explosion in application, particularly in areas such as network management, finance, medicine, and bioinformatics.

The characterization of machine intelligence given in the paragraph above differs to a large degree to the classification of machine learning systems that is given in this book, but there are similarities. Since the time of publication of this book, emphasis now is placed on the ability of machines to solve problems in more than one domain. In particular, many insist that a truly intelligent machine must be able to use essentially the same reasoning patterns over multiple domains, or as a bare minimum that a reasoning pattern be able solve a problem in one domain and without alteration solve a different problem in a different domain.

The editors of this book classify machine learning systems on the basis of the learning strategies that they use, the way that they represent the knowledge or skills that they acquire, and in terms of the domain in which the knowledge is acquired. It is interesting to note however that they distinguish the learning strategies by the amount of inference that the machine is able to perform. A machine that is programmed directly will not perform any inference, even though its knowledge increases. One could call these ordinary machines, and they represent most of the machines at the present time. However, there are at the present time machines that can discover new theories and invent new concepts. These machines the editors argue perform a substantial amount of inference, and therefore are less dependent on the need for tutoring or guidance by a teacher or the external environment. There are machines today that are certainly able to engage in this type of inference. Their reasoning patterns on based on inductive logic and they have been used in automated drug discovery. This book contains other examples of this type of learning, such as learning from examples and from instruction. The ability of some machines today to do this type of learning is a definite sign of the remarkable advances in machine intelligence since this volume was written.

This is a great book because it focuses on machine learning techniques. It has been used as textbook in my class.

Although this text book is not required in my data mining class, but I found it is very helpful for my study. I highly recommend this one to any new commer to the field of machine learning and data mining.

I am using this textbook for a Machine Learning class. While my professor is excellent, I must say that this book is a welcome addition to class. It is so well written that I actually enjoy reading it. The examples are well structured and it greatly helped me understand the material.

First of all, the statistical part of machine learning is JUST a real subset of mathematical statisitcs, whatever Bayesian or frequentist. Secondly, the non-statistical part of machine learning is not systematic or well developed yet.

We have used it as a textbook for a half-year course of Machine Learning. I think

[1] this book is a collection of many methods from Pattern Recognition, Statistics, Coding Theory, etc.
[2] anyone who wants to make clear the algorithms of EM, ME, etc, DO NOT read this book.
[3] it is suitable for undergraduate students in light of its simple mathematics.
[4] the examples in this book are very very naive.

I liked the book. But I think author must provide more figures in the book like Duda and Hart's Pattern Classification book. I used the book as master course and I found it easy to follow, interesting and useful book even I am newcomer to topic.

just the right content and texr easy to read!

It's pretty well done, it covers theory and core areas but - maybe it was more the state of the field when it was written - I found it unsatisfyingly un-synthesized, unconnected, and short of detail (but this is subjective). I found the 2nd edition of Russell and Norvig to be a better introduction where it covers the same topic, which it does for everything I can think of, except VC dimension.

The book sorely needs an update, it was written in 1997 and the field has moved fast. A comparison with Mitchell's current course (materials generously available online) shows that about 1/4 of the topics taught have arisen since the book was published; Boosting, Support Vector Machines and Hidden Markov Models to name the best-known. The book also does not cover statistical or data mining methods.

Despite the subjective complaint about lack of depth it does give the theoretical roots and many fundamental techniques decently and readably. For many purposes though it may have been superceded by R&N 2nd ed.

This textbook is useful too much for students that need to learn about learning algorithms. In this textbook explains the key theory and algorithms to solve various problems. For student who plan to make a research in machine learning, this textbooks can give basic knowledge and background of the research.

This book serves as an excellent introduction to machine learning. The material covers a broad range of important machine learning concepts and algorithms. The text is well-written and well-organized, and I use it frequently as a reference. In addition to describing the basic theory and mechanics of the algorithms, the book helps develop an intuitive feel for the algorithms and provides examples of how they have been used.

That said, the coverage of the theory behind the algorithms is fairly superficial. Additionally, having been written in 1997, many recent algorithms such as SVMs and methods such as Adaboost are not covered. For those reasons, this book cannot serve as a stand-alone text book for a course.

When I came to the field of machine learning, the book provides me a clear, easy-understanding picture to the field so that I believe any of you can get into the field by use of the book. If you are looking for your first book to this field, don't waste your time, it is. Even through my life in the research, I depended on it most of the time. It's too great.

This book is a good introduction to the field, but I think the notation can be quite cumbersome at times. I've seen the concepts presented elsewhere in less confusing form, but it's a good general source as it includes a considerable amount of information from relatively current research. The examples are typically very easy to understand, though they aren't always complicated enough to make the notation easy to understand.

Excellent book , state-of-the-art, nice presentation and covering lots of topics in a friendly manner. highly recommended !

I am a graduate student at a major research university. I am currently taking my fifth AI/Machine Learning graduate course. This is the one book everyone grabs for when they need a reference. I had to mark the spine of my book with tape so I could find it more easily on my colleagues shelves.

Other books are either not as accessible or too niche-specific. This is the only book out there that covers all of the major machine learning techniques (with the possible exception of support vector machines) and covers them in a manner that can be well understood.

Every discipline has one book that must be on your shelf. If you are planning on doing serious research in Machine Learning - this is the one book.

I got this book from the university's library, because I wanted a nice book that can show me different methods for machine learning, so I can learn the buzzwords, and understand their meaning, at least in principle. It was important to me that the book won't go too much in depth into any subject, and more importantly, that the book won't use unfamiliar terminology, unless explained before.

This book is indeed a nice overview of the field at the time it was written, although a lot happened in Machine Learning since, the book remains a good source for learning what can be done and how.

Reading the book is quite easy (I'm a graduate student in computer science), and quickly I got what I was asking for.

This book is not a magic pill for any problem. This is not a cookbook. So the compliments are in place as long as you know what you're looking for.

These days I attend a seminar where the students are asked, each student in turn to present one chapter from this book (and some other books). Surely, once the seminar is over, we (the students) will know enough to be able to chat about Machine Learning, and more importantly, we will know where to look for deeper texts if we wanted to.

I agree with some of the previous reviews which criticize the book for its lack of depth, but I believe this to be an asset rather than a liability given its target audience (seniors and beginning grad. students). The average college senior typically knows very little about subjects like neural networks, genetic algorithms, or Baysian networks, and this book goes a long way in demystifying these subjects in a very clear, concise, and understandable way. Moreover, the first-year grad. student who is interested in possibly doing research in this field needs more of an overview than to dive deeply into
one of the many branches which themselves have had entire books written about them. This is one of the few if only books where one will find diverse areas of learning (e.g. analytical, reinforcment, Bayesian, neural-network, genetic-algorithmic) all within the same cover.

But more than just an encyclopedic introduction, the author makes a number of connections between the different paradigms. For example, he explains that associated with each paradigm is the notion of an inductive-learning bias, i.e. the underlying assumptions that lend validity to a given learning approach. These end-of-chapter discussions on bias seem very interesting and unique to this book.

Finally, I used this book for part of the reading material for an intro. AI class, and received much positive feedback from the students, although some did find the presentation a bit too abstract for their undergraduate tastes

Mitchell provides a good coverage of ML subject matter but niether goes right in-depth, nor gives a digestable overview.

I'm a post grad student working in the area, and expect to find one of two things:
1. a clear & concise overview that can bring me up to speed on a body of research - what the competing theories/methods are, when they would be useful, a review of business cases or experimental results justifying various academic positions.
OR
2. In depth analysis of the theory and methods of implementation.

I never felt as if I got either from ML.

I teach AI at the graduate level in a major US research University, and I specialize in the area. The book does cover many different areas of Machine Learning. Unfortunately, the treatment is quite superficial. A student would find it extremely difficult to grasp imortant concepts without referring to other material. It may be a good reference, but I would definitely not recommend it as the main textbook. Unfortunately, there seem to be very few books in this area adequate for a senior or graduate level course.

If you are interested in the topic of machine learning, this is the start book for you. It is quite informative and covers the actual methods of learning and the concepts involved. It is not the most readable but the author does a good job livening it up. For that matter, I do not know how to convey the topic in a readable fashion so compared against other similar books, the presentation is outstanding. For those in the field, this is a must.

This book is completely worth the price, and worth the hardcover to take care of it. The main chapters of the book are independent, so you can read them in any order. The way it explains the different learning approaches is beautiful because: 1)it explains them nicely 2)it gives examples and 3)it presents pseudocode summaries of the algorithms. As a software developer, what else could I possibly ask for?

In fall 2000, I taught a master's level course in ML to about 25 students at New York University. Fortunately both for me and my students, I was able to use and assign excellent recent textbooks in the area: "Machine Learning" by Tom Mitchell and "Data Mining: Practical Machine Learning Tools and Techniques with Java Implementations" by Ian H. Witten and Eibe Frank. I recommend both books enthusiastically.

A student who has mastered Mitchell has a solid grasp of the basic element of nearly every method of machine learning currently in use, and of almost every aspect of ML research. A student who has mastered Witten/Frank has a deep knowledge of the major ML techniques, and a strong sense of the opportunities and pitfalls to be encounted when these techniques are put into practice....

One of the best books on the subject. Mitchell gives a good introductory coverage to all aspects of Machine Learning. This is not a book full of mathematics, it is a book that gets across ideas and concepts.

It is very rare to find a text that both does rigorous justice to a subject, and also is an enjoyable read. This book is such a rarity

I first used this book as the required text for my course in ML in 1997 and got rave reviews from the students. I will be using it again in 1999. I found ALL of the major topics and issues in ML addressed. The book is easily readable with anyone with a computer science background, and the book works quite well in a wide variety of approaches to presentation at the advanced undergraduate and graduate levels.

Everything I will do in the future will be based on ML and just one semester of an ML course & this book has converted me(even though my major is not Comp.Science). Of-course this is due majorly to Dr. Thomas Ioerger and his teaching abilities(Texas A&M), but the book presents all concepts(even seemingly complex ones) in a way that is easy and enjoyable to learn. One of the most useful books I've ever studied!

This book does an incredible job of presenting sophisticated material in a clear and easy to understand style. I highly recommend it to anyone interested in the field. Absolutely first rate!