Computing Reviews
Today's Issue Hot Topics Search Browse Recommended My Account Log In
Review Help
Search
Machine learning techniques for multimedia (1st ed.): case studies on organization and retrieval (Cognitive Technologies)
Cord M., Cunningham P., Cord M., Cunningham P., Springer-Verlag Telos, Santa Clara, CA, 2008. 290 pp. Type: Book (354075170X9783540751700)
Date Reviewed: Sep 8 2008

A rich collection of chapters dealing with various issues in the area of machine learning techniques used in multimedia information retrieval, this book covers a number of timely issues, with case studies on topics such as analyzing, classifying, and clustering image, video, and audio data. The chapters are organized in such a way that readers of various levels can quickly gain insight into multimedia retrieval using machine learning techniques. This is a very well-written book that can be used by practitioners and academics alike.

The book is organized into two major parts. The first part consists of four chapters, and presents an overview of various common machine learning techniques used in multimedia information retrieval. Chapter 1 discusses the basics of Bayesian methods and decision theory--the core of many machine learning algorithms. The subjects of supervised and unsupervised learning are covered in chapters 2 and 3. I was particularly delighted that the authors brought up the topic of dimension reduction in chapter 4. Dimension reduction is extremely important in modern information retrieval, because we are now working with massive amounts of data from the Internet that were not a concern of traditional information retrieval systems. Dimension reduction makes it possible to apply traditional information retrieval and various machine learning techniques to the data on the scale of the Internet. Readers who have not worked with machine learning before will be able to gain a basic understanding of the various techniques involved after reading this part of the book. Those who are fluent in machine learning techniques can skip these chapters.

The second part of the book concentrates on applying various machine learning techniques to multimedia information retrieval. Each chapter is a case study of a particular system. The chapters are organized in a uniform pattern, starting with the basic principles and followed by experiment setup and data analysis. This part has seven chapters, each presenting a specific multimedia information retrieval system that uses some machine learning techniques. Six of the seven chapters deal with still images or videos; the last chapter treats the subject of music classification and clustering.

Although various media--such as image, video, and audio--are used as application examples, different themes of machine learning and multimedia information retrieval are covered in various chapters. Chapter 5 is concerned with visual information retrieval systems based on supervised learning. Incremental learning in the context of video analysis is the subject of chapter 6. Chapter 7 considers several machine learning algorithms used for face recognition, and recommends Bayesian network classifiers for face detection and facial expression analysis. Explicit and implicit content query (query by example) are the subjects of chapter 8; the chapter compares the two with experiments and data. Chapter 9 investigates the issue of semantic labeling to aid image and video information retrieval. Chapter 10 presents the reader with the concept of a semi-structured way to handle multimedia information retrieval--that is, finding and constructing structures, both in content and in organization, with multimedia documents to help document filtering and categorization. While the first six chapters of Part 2 deal with image and video, the last chapter of the book features information retrieval with audio. The chapter discusses the basic idea of audio feature extraction and classification, followed by a couple of experimental music retrieval systems.

The book is very well written. It is accessible to a wide variety of readers because of its organization: it provides an overview of the mathematical tools needed, followed by a collection of case studies that involve sample multimedia information retrieval systems. Readers will find it both easy to understand and accurate. The case studies are sufficiently detailed so that readers will be able to follow them. This book is suitable for a graduate-level course or for practitioners in the field.

Reviewer:  Xiannong Meng Review #: CR136037 (0907-0636)
Bookmark and Share
  Featured Reviewer  
 
Information Search And Retrieval (H.3.3 )
 
 
Cognitive Simulation (I.2.0 ... )
 
 
General (I.2.0 )
 
 
Learning (I.2.6 )
 
 
Multimedia Information Systems (H.5.1 )
 
 
Sound And Music Computing (H.5.5 )
 
Would you recommend this review?
yes
no
Other reviews under "Information Search And Retrieval": Date
Nested transactions in a combined IRS-DBMS architecture
Schek H. (ed)  Research and development in information retrieval (, King’s College, Cambridge,701984. Type: Proceedings
Nov 1 1985
An integrated fact/document information system for office automation
Ozkarahan E., Can F. (ed) Information Technology Research Development Applications 3(3): 142-156, 1984. Type: Article
Oct 1 1985
Access methods for text
Faloutsos C. ACM Computing Surveys 17(1): 49-74, 1985. Type: Article
Jan 1 1986
more...

E-Mail This Printer-Friendly
Send Your Comments
Contact Us
Reproduction in whole or in part without permission is prohibited.   Copyright 1999-2024 ThinkLoud®
Terms of Use
| Privacy Policy