Computers and Intractability: A Guide to the Theory of NP-Completeness (Series of Books in the Mathematical Sciences) (Paperback)

by M. R. Garey (Author), D. S. Johnson (Author)

Editorial Reviews

Amazon.com Review
This book's introduction features a humorous story of a man with a line of people behind him, who explains to his boss, "I can't find an efficient algorithm, but neither can all these famous people." This man illustrates an important quality of a class of problems, namely, the NP-complete problems: if you can prove that a problem is in this class, then it has no known polynomial-time solution that is guaranteed to work in general. This quality implies that the problem is difficult to deal with in practice.

The focus of this book is to teach the reader how to identify, deal with, and understand the essence of NP-complete problems; Computers and Intractability does all of those things effectively. In a readable yet mathematically rigorous manner, the book covers topics such as how to prove that a given problem is NP-complete and how to cope with NP-complete problems. (There is even a chapter on advanced topics, with numerous references.) Computers and Intractability also contains a list of more than 300 problems--most of which are known to be NP-complete--with comments and references.

Product Details

• Paperback: 340 pages
• Publisher: W. H. Freeman (January 15, 1979)
• Language: English
• ISBN-10: 0716710455
• ISBN-13: 978-0716710455
• Product Dimensions: 9.1 x 6.5 x 0.7 inches
• Shipping Weight: 1.1 pounds (View shipping rates and policies)
• Average Customer Review: 4.8 out of 5 stars See all reviews (13 customer reviews)
• Amazon.com Sales Rank: #243,730 in Books (See Bestsellers in Books)

  Popular in this category: (What's this?)

  #5 in

  Books > Computers & Internet > Programming > APIs & Operating Environments > Device Drivers

Customer Reviews

Most Helpful Customer Reviews

30 of 31 people found the following review helpful:

5.0 out of 5 stars Contemplating Abstract Thought, April 29, 2000

By	S. Wuest (Tucson, AZ United States) - See all my reviews (REAL NAME)

Every graduate CS student will probably encounter this book--it is a classic.

But long after that course in NP theory was over, the astonishment of a different aspect of the book remains.

One course assignment was the development of 15 polynomial reduction proofs (proving the computational complexity equivalence of pairs of NP problems). Part of these proofs can be simple geometric shapes, locations and connecting lines, which are defined as elements in the 2 problems. Because the elements are rigorously defined, the resulting geometric pictures model rigorous proofs of equivalence.

I was astounded at the power of such abstractions (which most programmers perhaps wouldn't even recognize as legitimate proofs). This experience underlined the fact that rigorous logical proof may take many outer forms, whether mathematical equations, formal symbolic logic proofs such as the Irving Copi notation, or simple geometric drawings.

Many veins of rich ore may be mined from this work, and only 1 of them is NP theory. But the reader must be ready to do battle with some difficult ideas, and mathematical notation which can obscure the creativity of the material covered. (For astounding creativity, examine Cooke's Theorem proving that the Satisfiability problem is NP-Complete!)

16 of 16 people found the following review helpful:

5.0 out of 5 stars On the difficulty of computers to deal with certain problems, April 16, 1997

By A Customer

All those who deals with Computer Science, Mathematics and Engineering have to face the reality that certain problems seem really hard to solve. Even with the more sophisticated, and technologically advanced among the currently available computers---and among those that are to come in the next several years---, it seems highly likely that we cannot efficiently solve certain specific problems.

A first well written and systematic account on the hardness of this problems is the 1979 book on the theory of NP completeness by Michael R. Garey and David S. Johnson: Computers and Intractability, A Guide to the Theory of NP-Completeness (W.H. Freeman and Company, San Francisco). It is amazing how, after all these years, this book remains a fundamental one to be introduced on what can be effectively and efficiently solved by computers and above all on what it seems not efficiently solvable, independently of the advances of technology. Other texts have been published after that one, as for example the recent clear and complete overview on what has been done and extensively researched since then that has been given by Christos H. Papadimitriou in his book Computational Complexity (Addison-Wesley, 1994). Nevertheless, the Garey-Johnson book---as it is often familiarly called---remains the fundamental book for a clear introduction to this central problem of what is tractable by computers.

Starting from a very clear introduction to the technical term "NP-Complete," and to how this term gained importance for the description of the algorithmic tractability of certain problems in the early 70s, the book clearly defines, both in an intuitive and then in a formal way, what it is meant by the complexity of a problem. More than that, this complexity is directly related to the effective methods for solving problems (algorithms) and thus to computers themselves. The basic of the theory of NP-Completeness is completely covered in the first 5 chapters, beginning from a low-level introduction to some of the central notions of computational complexity and finally providing detailed definitions describing proof techniques to prove the hardness of certain problems. The remaining two chapters provide an overview on two alternative directions for further study. (The both of them have been extensively investigated in the following years.) Finally, the appendix contains more than 300 main entries on NP-Complete and NP-Hard problems, and this last part of the book is continuously referenced in journal and conference papers on the subject.

The first chapter is definitely accessible also to those that doesn't know so much mathematics, or computers related stuff, and thus the book is recommendable to those that are simply curious about the things that can be solved with computers.

16 of 19 people found the following review helpful:

4.0 out of 5 stars Showing its age, December 2, 2001

By A Customer

Yes, it's a classic. Yes, every computer scientist MUST own it. But enormous significant progress has been made in the field of NP-completeness (and computational complexity more generally) in the two decades since this book was published. An up-to-date edition -- which would probably be well over a thousand pages long -- has been badly needed for years.