Customer Reviews

Web Data Mining: Exploring Hyperlinks, Contents, and Usage Data (Data-Centric Systems and Applications)

5 star:		(2)
4 star:		(0)
3 star:		(1)
2 star:		(0)
1 star:		(0)

 

 

So what does the author, Bing Liu know about Web data mining to write the book "Web Data Mining - Exploring Hyperlinks, Contents, and Usage Data"[1] ? Fortunately the answer is "a lot!" This fact along with the title which had some cosine similarity with the names of my research lab and a graduate course that I have been teaching at the University of Louisville since 2004, and prior to that at the University of Memphis since 2000, are the reasons why I ordered a copy of this book. Bing Liu is a well seasoned researcher who has made significant contributions to association rule mining, in particular classification using association rule mining and association rule mining with multiple supports. He has also worked on Web data extraction, and more recently on opinion mining. In addition to the expertise of the author, two of the chapters, Chapter 8, Web Crawling, and Chapter 12, Web Usage Mining, were contributed by two leading experts in these respective areas, Filippo Menczer for the former and Bamshad Mobasher for the latter.
This book is appropriate for students at the graduate or senior undergraduate level, for practitioners in industry, and even as a good comprehensive reference for researchers in academia.
The Table of Contents held a surprise for someone who had always found it hard to limit the number of textbooks to one book in a web mining course that does not have data mining as prerequisite, and thus typically prescribes a good data mining book to introduce data mining techniques, in addition to a second book related to web mining. This book, on the other hand, has two parts, one devoted to data mining, and the other devoted to Web mining. While it was not a problem to find a very good data mining book (I have a few of them on my bookshelf), it was harder to find a book that addressed data mining and Web mining. It was also hard to find a good and comprehensive Web mining book, since most of them tend to focus on one or only two of the three main Web mining areas of Web structure, content, and usage mining (typically leaving Web usage mining in the dark, with just a small section, citing that it is an emerging area). This book, on the other hand, is a serious book on Web mining that also devotes a decent portion to data mining. I would describe the way the topics are presented as deep and rigorous enough in most chapters, which is in contrast to a large number of books on data mining and web mining. That said, because the book is full of simple examples that illustrate the methods being discussed, it is useful even for beginners, making it also appropriate for an introductory level course.

What I liked most about the book was the scratch I got when facing all the possibilities regarding data that is free available on the Internet. My interest area is crawling, and there is an exclusive chapter about it on the book. But as with all others chapters, it's only a bird's-eye view on the topic, so specifically the crawler part of the book wasn't of much use. In spite of it, my expectations were reached with the rest of the work, since I just wanted to be aware of what is happening today concerning Web data mining. I must note that, although chapters on relevant topics are small (more or less 30 and so pages) and surely don't cover all the nuances, the book comes with plenty of references for anyone who wants to dig further.

This book makes a great text for graduate courses, as well as a reference for scholars. The chapters are well written and provide good examples for any significant concepts. Each section covers the basics to establish a foundation of understanding for someone unfamiliar with the area, but goes on to also touch upon the research forefront on each topic. One of the most useful sections I've found as a researcher is the Bibliographic Notes found at the end of each section which briefly describes the major groups of work within the topic with cites to major papers/articles/books in each of these areas (seems to be about 50 or so per chapter).

The only "drawback" to this book would be if you wanted to touch upon everything, there is far too much content for a single semester. However as mentioned above, the chapters are structured such that you could easily use the first couple sections of each chapter to cover all the foundations and either leave later sections for students to read on their own/select an advanced project, or cover the remainder in a 2nd semester.

I highly recommend this book to any graduate looking for a comprehensive text and reference on web mining.

(In the interest of full disclosure, I am listed in the acknowledgements from providing feedback on a pre-print edition of the text that was used as our course textbook. I do not get royalties from sales in any way.)