Browse Books

Reviewer: Claudiu Popescu

This volume contains eight research papers on the relation between Web 2.0 and the semantic Web. The book has an advanced theoretical approach. It is addressed to specialists in Web development, but it will also benefit anyone interested in recent Web 2.0 developments. While in the classical Web, the user is only a reader of Web pages, in Web 2.0, the user is also able to create data-for example, the user can post pictures with Flickr, create Facebook accounts, and make comments. In the classical Web, a page can only be accessed by its uniform resource locator (URL), but in the semantic Web, the pages also have tags and annotations, so that the data can be accessed by its meaning (semantics). The papers in this volume describe new technologies that are applicable to both Web 2.0 and the semantic Web. Each paper includes formal definitions and many references. Although the theory is advanced, a broader audience will find the introductions and conclusions manageable. Since the papers cover different areas, this review briefly summarizes their content. The first paper presents the TagFusion system. Due to the exponential increase in the amount of information published on the Web, plain browsing is no longer adequate. There is a need for semantic metadata-for example, tags associated with data-to identify the content of the data. In the tagging process, either the user manually assigns words to the published content or the tags are automatically collected from the text. Tags are easy to assign, they can even be new words, and they are useful for classification. On the other hand, tags can be ambiguous-one tag can mean different things or two tags can be synonymous-and can lack semantics-for example, the tag "music" may refer to the text of an audio file. Folksonomy, an extension to the tagging system, is where a tag contains three entities: the keyword, the user who assigned the tag, and the object tagged. The rest of the paper presents TagFusion, the authors' system. TagFusion is a tagging system that can be shared between applications that normally have proprietary tagging systems, such as Flickr and Technorati. A second paper on tagging describes GroupMe!, a system that allows the user to create a group and drag and drop various Web resources into it. Then, by learning the relation between the elements dragged into the group, the application generates new semantics. A mathematical definition of a folksonomy is presented in the authors' theoretical description of GroupMe! The third paper discusses adaptation techniques for improving annotations (advanced tagging). Such improvements include minimizing ambiguity and using multifaceted annotations that provide information about the tag. The fourth paper describes a very complex system that adapts Ajax-based rich Internet applications (RIAs). The system changes the user interface, based on a set of rules involving semantic Web usage patterns and page annotations. The fifth paper is titled "Towards Enhanced Usability of Natural Language Interfaces to Knowledge Bases," and discusses finding information on the Web using a natural language. The sixth paper introduces the semantic document model, which is useful for creating a desktop with uniform access to all resources, including Word documents, PDFs, and PowerPoint files. It also allows for the interoperability of knowledge bases. The seventh paper describes annotation-based data mining in digital libraries. This allows for the identification of duplicated annotations (for example, upper- and lower-case letters and abbreviations), matching titles, and cleaning the database. The last paper, "An Assessment System on the Semantic Web," discusses a Web application to test students and then process the data. The system described is based on the question and test interoperability (QTI) standard. Most of the paper shows the implementation of the system using the Eclipse modeling framework (EMF). However, this paper does not seem to be related to the semantic Web, other than the title. For advanced researchers, this useful book covers many new topics and presents lots of technical solutions. Online Computing Reviews Service