Computer Algorithms

There are essentially three reasons one might want a book on algorithms: (a) to learn how to design algorithms; (b) to learn how to analyze algorithms; and (c) as a reference for algorithms that are already known. This is a decent book for the first purpose, and a poor choice for the remaining two. Organizationally, it is structured around general techniques for algorithm design, which means that techniques of analysis and algorithms for solving specific problems are scattered widely and difficult to find or link together into a logical sequence. I also find myself sympathetic with other reviewers who complain about the difficulty in understanding the book. The problem is not that the information is difficult, but that the authors often do not approach it seriously. It's easy to tell that they are interested in the general techniques, and that they considered it a chore to explain the mathematics behind analysis. As a result, the mathematical side of the book is not particularly inspired. That said, though, this is the ONLY book I'm aware of that teaches with this particular organization. If your goal is to create new algorithms for solving problems (particularly combinatorial problems), then this is probably worth a read even if you have to refer elsewhere to fill in some details.

I dont know why this books has such negative reviews. Its one of the only algo books that covers so many topics in detail.In most Indian universities, "Introduction To algorithms" By Comer(MIT press) is the standard book for undergrad and Knuths book is used for Grad. But the Comer book is too theoretical. It does not tell you how to think. The Horowitz/Sahni book gives you an idea of how to think intuitively. To see what I mean, read "Greedy algorithms" chapter from both the books and you will start loving Horowitz/Sahni book. My advice is that first understand the concepts from Horowitz/Sahni - it will also cover most maths. Then read Comer. And dont skip the problemsin Comer.

Very comprehensive in algorithms. Divede-and-conquer, greedy methods, dynamic programming, search techniques, branch-and-bound, NP-complete, and other important and advanced topics that an algorithm course might have. It's very good to use it as a text book.

Exactly I want to see some important pages of this book

I had the privilege of having Mr. Sahni as my data structures professor at the Univ. of Florida a few years back. This book co-authored by him is a valuable resource for the computer science student and professional as well. Updated in 1997, it's the best and most current source I've seen for pure algorithms and data structures. Highly reccomended.