Modern Information Retrieval

by Baeza-Yates, Ricardo; Ribeiro-Neto, Berthier

• ISBN13: 9780201398298
• ISBN10: 020139829X
• Format: Paperback
• Copyright: 1999-01-01
• Publisher: Addison Wesley

Summary

Information Retrieval/Database Management Modern Information Retrieval Ricardo Baeza-Yates and Berthier Ribeiro-Neto We live in the information age, where swift access to relevant information in whatever form or medium can dictate the success or failure of businesses or individuals. The timely provision of relevant information with minimal 'noise' is critical to modern society and this is what information retrieval (IR) is all about. It is a dynamic subject, with current changes driven by the expansion of the World Wide Web, the advent of modern and inexpensive graphical user interfaces and the development of reliable and low-cost mass storage devices. Modern Information Retrieval discusses all these changes in great detail and can be used for a first course on IR as well as graduate courses on the topic. The organization of the book, which includes a comprehensive glossary, allows the reader to either obtain a broad overview or detailed knowledge of all the key topics in modern IR. The heart of the book is the nine chapters written by Baeza-Yates and Ribeiro-Neto, two leading exponents in the field. For those wishing to delve deeper into key areas there are further state-of-the-art chapters on specialized topics, fully integrated and using the same notation and terminology, written by leading researchers: Parallel and Distributed IR - algorithms and architectures (Eric Brown) User Interfaces and Visualization - the main interface paradigms for query formation and visualization of results (Marti A. Hearst) Multimedia IR: Models and Languages - including MULTOS and SQL3 (Elisa Bertino, Barbara Catania and Elena Ferrari) Multimedia IR: Indexing and Searching - R-trees and GEMINI and QBIC (Christos Faloutsos) Libraries and Bibliographical Systems - online systems and public access catalogs (Edie M. Rasmussen) Digital Libraries - the main challenges for effective deployment (Edward A Fox and Ohm Sornil) In addition, the book contains several chapters on Text IR - all the main IR models, query operations, text operations, indexing and searching (three of them co-authored with Gonzalo Navarro or Nivio Ziviani) The Web - challenges, measures and models, search engines, directories, query languages, metasearches and trends Modern Information Retrieval will prove invaluable to students in Computer Science, Information Science and Library Science departments, as well as to programmers and analysts working with products related to the Web, intranets, document database systems and digital libraries. Visit Addison Wesley Longman on the World Wide Web at: http://www.awl-he.com/computing/ http://www.awl.com/cseng/

Table of Contents

Preface	v
Acknowledgements	vii
Biographies	xvii
Introduction	1	(18)
Motivation	1	(2)
Information versus Data Retrieval	1	(1)
Information Retrieval at the Center of the Stage	2	(1)
Focus of the Book	3	(1)
Basic Concepts	3	(3)
The User Task	4	(1)
Logical View of the Documents	5	(1)
Past, Present, and Future	6	(3)
Early Developments	6	(1)
Information Retrieval in the Library	7	(1)
The Web and Digital Libraries	7	(1)
Practical Issues	8	(1)
The Retrieval Process	9	(1)
Organization of the Book	10	(5)
Book Topics	11	(1)
Book Chapters	12	(3)
How to Use this Book	15	(2)
Teaching Suggestions	15	(1)
The Book's Web Page	16	(1)
Bibliograph Discussion	17	(2)
Modeling	19	(54)
Introduction	19	(1)
A Taxonomy of Information Retrieval Models	20	(1)
Retrieval: Ad hoc and Filtering	21	(2)
A Formal Characterization of IR Models	23	(1)
Classic Information Retrieval	24	(10)
Basic Concepts	24	(1)
Boolean Model	25	(2)
Vector Model	27	(3)
Probabilistic Model	30	(4)
Brief Comparison of Classic Models	34	(1)
Alternative Set Theoretic Models	34	(7)
Fuzzy Set Model	34	(4)
Extended Boolean Model	38	(3)
Alternative Algebraic Models	41	(7)
Generalized Vector Space Model	41	(3)
Latent Semantic Indexing Model	44	(2)
Neural Network Model	46	(2)
Alternative Probabilistic Models	48	(13)
Bayesian Networks	48	(1)
Inference Network Model	49	(7)
Belief Network Model	56	(3)
Comparison of Bayesian Network Models	59	(1)
Computational Costs of Bayesian Networks	60	(1)
The Impact of Bayesian Network Models	61	(1)
Structured Text Retrieval Models	61	(4)
Model Based on Non-Overlapping Lists	62	(1)
Model Based on Proximal Nodes	63	(2)
Models for Browsing	65	(4)
Flat Browsing	65	(1)
Structure Guided Browsing	66	(1)
The Hypertext Model	66	(3)
Trends and Research Issues	69	(1)
Bibliographic Discussion	69	(4)
Retrieval Evaluation	73	(26)
Introduction	73	(1)
Retrieval Performance Evaluation	74	(10)
Recall and Precision	75	(7)
Alternative Measures	82	(2)
Reference Collections	84	(12)
The TREC Collection	84	(7)
The CACM and ISI Collections	91	(3)
The Cystic Fibrosis Collection	94	(2)
Trends and Research Issues	96	(1)
Bibliographic Discussion	96	(3)
Query Languages	99	(18)
Introduction	99	(1)
Keyword-Based Querying	100	(4)
Single-World Queries	100	(1)
Context Queries	101	(1)
Boolean Queries	102	(1)
Natural Language	103	(1)
Pattern Matching	104	(2)
Structural Queries	106	(7)
Fixed Structure	108	(1)
Hypertext	108	(1)
Hierarchical Structure	109	(4)

Excerpts

Information retrieval (IR) has changed considerably in recent years with the expansion of the World Wide Web and the advent of modern and inexpensive graphical user interfaces and mass storage devices. As a result., traditional IR textbooks have become quite out of date and this has led to the introduction of new IR books. Nevertheless, we believe that there is still great need for a book that approaches the field in a rigorous and complete way from a computer-science perspective (as opposed to a user-centered perspective). This book is an effort to partially fulfill this gap and should be useful for a first course on information retrieval as well as for a graduate course on the topic. The book comprises two portions which complement and balance each other. The core portion includes nine chapters authored or co-authored by the designers of the book. The second portion, which is fully integrated with the first, is formed by six state-of-the-art chapters written by leading researchers in their fields. The same notation and glossary are used in all the chapters. Thus, despite the fact that several people have contributed to the text, this book is really much more a textbook than an edited collection of chapters written by separate authors. Furthermore, unlike a collection of chapters, we have carefully designed the contents and organization of the book to present a cohesive view of all the important aspects of modern information retrieval. From IR models to indexing text, from IR visual tools and interfaces to the Web, from IR. multimedia to digital libraries, the book provides both breadth of coverage and richness of detail. It is our hope that, given the now clear relevance and significance of information retrieval to modern society. the book will contribute to further disseminate the study of the discipline at information science, computer science, and library science departments throughout the world. Ricardo Baeza-Yates, Santiago, Chile Berthier Ribeiro-Neto. Belo Horizonte, Brazil January, 1999 020139829XP04062001