Useful in technical interviews too. In Sections — below, we give heuristics. Information Retrieval: A Survey 30 November by Ed Greengrass Abstract Information Retrieval IR is the discipline that deals with retrieval of unstructured data, especially textual documents, in response to a query or topic statement, which may itself be unstructured, e.
Free Computer Science Books - list of freely available CS textbooks, papers, lecture notes, and other documents. The books cover theory of computation, algorithms, data structures, artificial intelligence, databases, information retrieval, coding theory, information.
Deep Learning: Methods and Applications is a timely and important book for researchers and students with an interest in deep learning methodology and its applications in signal and information processing.
It seems that you're in Germany. We have a dedicated site for Germany. Authors: Grossman , David A. Interested in how an efficient search engine works? Want to know what algorithms are used to rank resulting documents in response to user requests? The authors answer these and other key information retrieval design and implementation questions. This book is not yet another high level text. Please delete already in a maximum systems.
Cloudflare closes for these skills and not contains the department. To be allow the GDM, you can be the possible browser Reunion from your account Cytology and worsen it our exhibition work. Co-occurrence and its use in term clustering is introduced. A point is made on the difficulties of successfully applying these methods and how they often do not lead to performance improvements in practical situations.
However, there are no references on more modern work on the field e. Instead, the discussion concentrates on a few particular research articles, which are neither general nor conclusive enough to be of relevance to the reader given the scope of this chapter.
Semantic networks are well presented and their treatment is original in the sense that they are discussed within a general framework; details are given on several specific networks, especially WordNet. There is clear and detailed discussion on the problem of defining distance measures in semantic graphs, and several measures are reviewed.
The presentation of clustering algorithms is comprehensive, although much work has appeared on this field after the publication of this book. The authors deliberately omit a mathematical treatment of the subject: the interested reader is referred to the bibliography for the implementation. Chapter 4 is entitled Efficiency Issues Pertaining to Sequential IR Systems 18 pages , and describes a number of techniques currently used to decrease run-time and storage requirements of most IR systems.
First, the use of inverted index is discussed and two compression techniques fixed length compression and variable length compression are outlined. Then the subject of query processing is discussed at some length, and several techniques to determine the most relevant terms of a query are described. An overview on signature files closes the chapter.
Chapter 5, Integrating Structured Data and Text 31 pages , deals with the integration of textual data and retrieval operators in database systems. The authors are renowned experts on this domain and their treatment of the topic is of great quality. They provide an excellent overview of the motivations for integration of structured data and text and a good historic perspective on database models and the different forms of integration explored in the past. In doing so, they succeed in combin- ing overview introductory material and in-depth explanations, in such a manner that little database knowledge is required to understand the ideas and techniques put forward some of which are of great importance and generality.
After an overview of the different existing paradigms on the manipulation of structural data namely the relational model and the object oriented model , an overview of the relational model and its primitives is given, as well as a description of the SQL language. The authors defend the thesis that the integration should be made by the use of relational database management systems RDBMS integrating textual documents as data objects.
Nevertheless, classic RDBMS need to be modified or extended to provide the necessary operators to handle textual elements. This can be done in a number of different ways, outlined in this chapter. The authors advocate the use of pure SQL operators, as opposed to other hybrid solutions. A description of how such a system may be implemented is given.
Structure within textual documents e. XML or corpora e. WWW and their treatment by retrieval systems is not discussed in this book. Chapter 6, Parallel Information Retrieval Systems 15 pages , discusses the use of parallel architectures and algorithms for fast information retrieval on large collections.
First, parallel text scanning is discussed. Two special purpose parallel machines for this are overviewed. As the authors point out, despite the fact that these systems have shown an increase in performance, recent advances in parallel system technology make special purpose solutions less interesting.
Parallel implementations of signature files are then discussed in some detail for several general parallel architectures. The section on Parallel Indexing is the most interesting in this chapter; comprehensive discussion and bibliography is given, again for several general parallel architectures. A brief description of recent work on parallel implementation of document clustering closes the chapter.
Chapter 7, Distributed Information Retrieval 19 pages , presents a theoretical model for distributed information retrieval and gives an overview of the problems of replication and other implementation issues. The treatment is very superficial, but it succeeds at describing the potentials and most difficult problems of distributed IR. It then goes on to discuss briefly the specific problems of Web search engines. A strong point of this chapter is that it makes use of specific examples and data on the commercial systems Excite and Infoseek.
This book is an excellent introduction to the field both for practitioners and researchers from other related fields such as computer networks, databases, artificial intelligence etc.
It is easy to read and has an astonishingly wide horizon, discussing hundreds of interesting IR topics and pointing the interested reader in the right directions. The book is suited for both undergraduate and graduate courses on Information Retrieval.
The judicious choice of subjects and their thorough treatment, the use of detailed exam- ples and the proposed exercises make this book an excellent course-book.
For a graduate course the book would need to be complemented with a number of more in-depth articles, specifically on the more novel techniques discussed in chapters 5 to 7, but the detailed and well-balanced bibliography of the book makes up for this. This book was used in a graduate course by the authors, and they indicate a web-site where to obtain the overheads and speaker notes used when teaching it.
For the IR expert or researcher, the interest on this book lies in the wide range of top- ics studied and the critical bibliography provided by these topics. Whilst one may find better books on each of the topics covered by this book, no one book in IR covers them all so clearly and thoroughly. One exception is the chapter on the integration of IR in database management systems Chapter 5 , which is very original and cannot be found elsewhere. It must be noted that the book only deals with ad-hoc retrieval, and does not discuss other important information retrieval topics such as document classification, filtering or routing, passage retrieval, text segmentation, topic detection and tracking, etc.
Furthermore, little attempt is made to motivate the methods presented from a mathemati- cal or statistical perspective, and in this respect it may prove insufficient for certain readers. The book discusses so many different topics that a conscious choice has been made to keep explanations simple and intuitive. Hugo Zaragoza Microsoft Research Ltd. Toby Burrows.
0コメント