I am not sure how this book gets consistently high marks. I am using this text for a graduate level course. While it does a decent job covering most of the topics, it has some glaring flaws.

For one the Homework Problems it provides are not really representative of what you're learning in the text. Almost all of the problems revolve around proofs, as opposed to using the concepts in practice. You can seemingly have a good grasp on the material, yet spend hours trying to solve each of the problems they provide for that particular section. My entire class has complained, and even my professor has admitted that even he isn't sure sometimes how they expect you to solve some of the problems.

Secondly, there are very few example problems demonstrated in the text, so the reader doesn't really get to see the concepts in action so to speak.

Also, there is a typo or error on almost every other page, sometimes even on important formulas.

Overall, I'd have to think there are better books out there. If this truly is "the best there is" as some reviewers claim, God help the field of Pattern Recognition.

The 1973 book by Duda and Hart was a classic. It surveyed the literature on pattern classification and scene analysis and provided the practitioner with wonderful insight and exposition of the subject. In the intervening 28 years the field has exploded and there has been an enormous increase in technical approaches and applications.
With this in mind the authors and their new coauthor David Stork go about the task of providing a revision. True to the goals of the original the authors undertake to describe pattern recognition under a variety of topics and with several available methods to cover each topic. Important new areas are covered and old but now deemed less significant are dropped. Advances in statistical computing and computing in general also dictate the topics. So although the authors are the same and the title is almost the same (note that scene analysis is dropped from the title) it is more like an entirely new book on the subject rthan a revision of the old. For a revision, I would expect to see mostly the same chapters with the same titles and only a few new chapters along with expansion of old chapters.

Although I view this as a new book, that is not necessarily bad. In fact it may be viewed as a strength of the book. It maintains the style and clarity of the original that we all loved but represents the state-of-the-art in pattern recognition at the beginning of the 21st Century.

The original had some very nice pictures. I liked some of them so much that I used them with permission in the section on classification error rate estimation in my bootstrap book. This edition goes much further with beautiful graphics including many nice three-dimensional color pictures like the one on the cover page.

The standard classical material is covered in the first five chapters with new material included (e.g. the EM algorithm and hidden markov models in Chapter 3). Chapter 6 covers multilayer neural networks (a totally new area). Nonmetric methods including decision trees and the CART methodology are covered in Chapter 8. Each chapter has a large number of relevant references and many homework exercises and computer exercises.

Chapter 9 is "Algorithm-Independent Machine Learning" and it includes the wonderful "No Free Lunch" theorem (Theorem 9.1), a discussion of the minimum desciption length principle, overfitting issues and Occam's razor, bias - variance tradeoffs,resampling method for estimation and classifier evaluation, and ideas about combining classifiers.

Chapter 10 is on unsurpervised learning and clustering. In addition to the traditional techniques covered in the first edition the authors include the many advances in mixture models.

I was particularly interested in that part of Chapter 9. There is good coverage of the topics and they provide a number of good references. However, I was a bit disappointed with the cursory treatment of bootstrap estimation of classification accuracy (section 9.6.3 on pages 485 - 486). I particularly disagree with the simplistic statement "In practice, the high computational complexity of bootstrap estimation of classifier accuracy is rarely worth possible improvements in that estimate (Section 9.5.1)". On the other hand, the book is one of the first to cover the newer and also promising resampling approaches called "Bagging" and "Boosting" that these authors seem to favor.

Davison and Hinkley's bootstrap text is mentioned for its practical applications and guidance for bootstrapping. The authors overlook Shao and Tu which offers more in the way of guidance. Also my book provides some guidance for error rate estimation but is overlooked.

My book also illustrate the limitations of the bootstrap. Phil Good's book provides guidance and is mentioned by the authors. But his book is very superficial and overgeneralized with respect to guiding practitioners. For these reasons I held back my enthusiasm and only gave this text four stars.

I used the first edition of this book in a class on pattern recognition back in 1998. That old first edition did a great job of explaining the different aspects of pattern recognition as they were generally taught when the first edition came out in 1969. However, over the next 30 years the field expanded enough that a second edition was required. I purchased it, expecting an expanded version that went over the details as well as the first edition, and boy was I wrong. This second edition just glosses over the details of modern pattern classification techniques and doesn't show sufficient examples or even motivation for you to "get it". It's almost like the entire book is an introduction. I'm accustomed to the first chapter of a technical book being an overview that doesn't tell you much, but not the entire book. The only thing the second edition has to offer are slicker illustrations. My advice is find a copy of the first edition. It is very well put together. If you need additional material on subjects the first edition doesn't cover well, then go find more modern books specifically on those subjects. You may spend more money but at least you'll learn something.

This was my first purchase from amazon and I was totally impressed by the quality of the product and the service! I would buy again from the same seller and recommend others to do the same.

I have now used this book 3 times for a class. While the 1st edition did a nice job of covering the material in its time, the additions to in the 2nd addition are a disaster. What the book has going for it is that it at least lists the necessary material for such a course in the table of contents. However, all the additional material is poorly explained at best. The problem sets are too few and the ones that are included are generally weak.

I have tried to use this book, but after constant student complaints and my own difficulty with the text, I have finally concluded that the problem lies with the text and not with the users.

I think an indicator of problems was the large number of errors in the first printing; large here is an understatement. Even in later additions, the 4th, the size of the errata is huge. I think this is indicative of the authors' attention to detail and seriousness in preparation. I have found similar errors and ambiguities in the associate Computer Manual.

The bottom line is that this book has seen its final appearance in our curriculum. I would use any other text, even an older one.

There is simply not enough room or time to point out all the problems with this text. Do yourself a favor if considering this text for a class. Don't bother.

I have now used this book 3 times for a class. While the 1st edition did a nice job of covering the material in its time, the additions to in the 2nd addition are a disaster. What the book has going for it is that it at least lists the necessary material for such a course in the table of contents. However, all the additional material is poorly explained at best. The problem sets are too few and the ones that are included are generally weak.

I have tried to use this book, but after constant student complaints and my own difficulty with the text, I have finally concluded that the problem lies with the text and not with the users.

I think an indicator of problems was the large number of errors in the first printing; large here is an understatement. Even in later additions, the 4th, the size of the errata is huge. I think this is indicative of the authors' attention to detail and seriousness in preparation. I have found similar errors and ambiguities in the associate Computer Manual.

The bottom line is that this book has seen its final appearance in our curriculum. I would use any other text, even an older one.

There is simply not enough room or time to point out all the problems with this text. Do yourself a favor if considering this text for a class. Don't bother.

The discussed book is very explanatory and could be students' material for academic lessons.

easy to read for computer scientists who are not necessarily experts in statistics. the code in matlab is very good, and helps a lot.
this book is a good introduction to machine learning.

I liked this book because it does a great job explaining the concepts and the reasoning behind the mathematical formulae. Other books such as "The Elements of Statistical Learning" toss the Math formulas at you and expect you to figure out the significance or the importance of 'em. The book does not shy away from Math - but does a great job presenting it.

Duda's 'Pattern Classification' is an excellent book for an introduction to a range of the most powerful and popular machine learning methods. Obviously it doesn't go into very many details of each method - being just one book - but it covers the most important things to know imo. Not quite the 20% that covers 80% of the needs, but close. It is well written and the illustrations are generally of high quality. If you don't want to drown in a nightmare of details and exceptions, but rather get a good basic understanding of the various techniques, this book is an excellent buy.

[1] This book is good as an introduction to Pattern Recognition, at undergraduate level (compared with the level of Fukunaga's -Introduction to Statistical Pattern Recognition-).
[2] The references may be helpful to those who are interested in kernel methods, SVM, etc for detailed discussions.
[3] Compared with Bishop's or Ripley's book on pattern recognition, this book is not a bible.
[4] As a textbook for undergraduate students, I will mark it 5 stars; as a reference book for researchers, it is worthy 3 stars at most.

Overall this is a good book on the field. There is plenty of examples and covers alot of topics (from Bayesian estimation, Support Vector Machines, LDA, PCA, Neural Networks etc.). My only disappointment with this book was in it's coverage of Hidden Markov Models. The algorith it presents is very confusing, and you are better off reading the Rabiner tutorial on the subject if you want to learn more or implement HMMs.

I found book very useful. Figures, mathematical explanations and algortihms provides complemantry information to understand topic better. There may some errors in the book but I did not found any fatal one. Problem questions of each chapter are very useful. This is a must book whom are interesting in pattern classification area both in industry and academy.

I found this book quite useful as an augmentive text to Elements of Statistical Learning used in a grad engineering level data mining course. This book is written more at an engineering level, and I found it to bridge well between advanced texts such as Elements of Statistical Learning and more general audience books that really are lacking. Duda and Hart do a good job at explaining the concepts, however some techniques only recieve a cursory overview while other topics are rather elaborated upon, however this may have been done by the authors experience of which techniques are commonly employed in practice. The excercises at the end of the chapters include a lot of hands on programming and computer-based assignments which I found useful, and a MATLAB workbook associated with this is also offered, however I have not read this book. Nonetheless I have implemented some of the concepts in this book using Matlab and it definately does help to cement the idea, even if this is just serves as an intellectual excercise and isn't intended to be used for anything else. With a little bit of digging through the help or using a book such as Ripley and Venerable's Modern Applied Statistics with S, most if not all of the techniques can be explored using the R statistical software.

If you think that some method such as SVM is the "holy grail" of machine learning and pattern recognition and are interested only in an in-depth coverage of that specific tool, this book is not for you. If, however, you want to understand the basic concepts and methods employed by a broad range of researchers and scientists, I highly recommend buying it.

The book covers a broad range of topics in pattern recognition. Its explanations are lucid, and its illustrations are helpful. The book is well-written and well-organized. When using this book as part of a low-level graduate course, I was not particularly impressed. Recently, however, I have found myself frequently going back to the book to refresh my understanding of the basic idea of some topic. I recommend PC as a companion text for a course in pattern recognition. I also recommend purchasing the book for private use.

This book is full of useful algorithms, as well as the theory behind them. The explanations are good, although they sometimes require reading them several times to fully grip what is going on and why it works (but this is the case with many useful algorithms). This is a must-have for pattern rec work.

Everything You ever want to know about the subject for intermediate .
Handbook of industries methods and algorithms for advanced .
Together with Mitchell perfect for beginners .

This book is not for the novice, and it assumes some mathematical skills on the reader's side.

Having read the book a few times now, I must conclude that this one covers a lot of ground regarding pattern classification, and is probably more complete than any other book currently on the market.

If you're really interested in pattern recognition, you will get through this book with success, and will feel very thankful about the many useful algorithms, all perfectly clarified with pictures and even pseudo code. For those having mathematical problems, you might have to read more than once or twice to get a good grip on it. Yes, there are a few bugs in there, but this is the same with anyother book (there is an errata available on the web).

I've also been implementing many of the algorithms discussed, and I think anybody seriously involved with pattern classification should have at least a copy of this book nearby.

For those who complain that the book doesn't cover enough topics, related to distributed processing, machine learning, statistical inference, etc, I think these topics don't belong in here (they deal less with pattern classification), and others have dedicated separate books for all of those topics.

For those readers/students complaining about "too complicated", or "too many errors", or "hard-to-understand concepts", I recommand a better science teacher.

I haven't read the first edition, but having been in the commercial field of data mining / data fore casting / data clustering for many years now, I think this book is very up-to-date. I have come to understand what works and what doesn't, and yes, maybe not everything is covered, but the things that are covered are definitely current and leading-edge technology.

I am using this book for class right now. Our professor complains about the book constantly because 1) the text is explained in too complicated of a way, 2) there are too many errors, and 3) some of the errors are quite mathematical in nature. Our professor said he tried to E-mail the author, but the author said he "didn't have time because so many people like the book."