This book addresses challenging issues in the new generation of web search, which aims to enhance traditional approaches by considering the semantics of a search query. Over the last 20 years, the web has become a major source of information for organizations and individuals. Information queries are enabled by web search, a mechanism for finding and retrieving information most relevant to the initiating search, whether that is a direct search entered by a human or a search operated by a machine for exchanging relevant information with other systems or humans. Web search has traditionally been based on extracting keywords from the text and evaluating their textual and structural similarity to the target content. Even though this approach has proven successful (we all use it every day, provided by our favorite search engine), it has been heavily criticized, mainly for its inability to accommodate the information needs of the seeker. This is where semantic search over the web comes in. Semantic search enables a system to interpret the meaning of the words in the query, or at least to simulate this understanding using synonyms and related terms, which produces additional information related to the information goal.
Semantic search is gradually becoming a reality. Major search engines already incorporate some semantic elements. However, there are still many challenges, the most serious of which relate to the size of the web, which exceeds the capacity of commonly used methods and tools. Thus, developing smarter methods for the search of relevant information is not the only important aspect. Without considering the scalability of techniques designed to discover, extract, and integrate data available on the web, we cannot move further. This book is the only one I have encountered that is deeply centered on data--or, more precisely, on effective and scalable methods and techniques for accessing information from the web, the world’s largest and most volatile database.
The editors managed to get a team of 57 contributors from 12 (mostly European) countries to write 14 chapters that together give an overview of recent research on semantic search and related areas (mostly in the field of databases and services). It is not easy to produce well-edited books, given the need to bring together the work of many authors, each with their own view on the subject. The editors of this book have done a good job of covering interesting topics and making the book readable as a whole. The chapters present deep insight into the research results of top teams, as evidenced by several well-known names among the contributors.
Even though the publisher’s web page (http://www.springer.com/computer/database+management+%26+information+retrieval/book/978-3-642-25007-1) advertises the book as the “first survey text on a hot new topic,” I do not consider it primarily as a survey. It goes deeply into several different aspects of the topic, which as a researcher I really appreciate. Each chapter is self-contained, with an introduction, conclusion, and references. Most chapters provide a motivating example.
The 14 chapters are divided into three parts. The first part introduces the notion of the web of data, and describes different formats of structured data published on the web. These include microformats, the resource description framework in attributes (RDFa), microdata and linked data; storing and indexing techniques for resource description frameworks (RDFs) as the basic formalism for modeling information on the web; and methods for accessing web data structures. Already this early in the book, the authors show that they understand the size issues on the web and describe techniques for storing and indexing massive RDF datasets.
The second part of the book includes eight chapters that offer deep insight into various aspects of semantic web search, including path-oriented search, keyword-based search, and ontological query answering. Two chapters present topics on queries, such as interactive query construction and answering a query without accessing instances (that is, without the possibility of prior access to data to construct an index). Another hot topic in semantic web search involves the linking of web data, considered a major challenge during semantic query construction. Linking is the topic of two chapters, one on semantic link discovery over relational data and the other on uncertainty in entity linking. This part of the book concludes with a discussion of web services and semantics-enabled mashups of linked data that support the composition of web application programming interfaces (APIs) for building applications.
Linked data appears in some form in almost every chapter. However, the third part specifically focuses on linked data and its use in search engines. This section includes a discussion of a recommender system, an application that discovers, extracts, and integrates data from web pages, and a chapter on a semantic web search engine that uses linked data to scale access to large amounts of data.
The book is primarily intended for graduate students and researchers in the field. Web developers interested in new technologies and deep theoretical insights will also appreciate the book. However, readers who lack experience with relational databases and RDF as the basic format for semantic representation will likely need to study supplemental material. However, for motivated novices in information systems or graduate computer science majors, the book will provide useful knowledge, inspirational ideas, and supporting references to other works.
