Handbook of Metadata, Semantics and Ontologies

The following sections are included:

• Preface
• Contents

The growth of the Web represents also one of its major challenges, as users face the problem of selecting the pages that are most relevant to their taskfrom a vast amount of information. Search engines and microdata are an example of a means towards the end of helping in better targeting search, but there is no universal perfect solution for all information needs. In addition, the last years have witnessed the emergence of the Web of Linked Data, fostered by the increasing adoption of openness as a paradigm for sharing information for the benefit of the commons. Microdata, linked data and other technologies are no other thing than different ways of using metadata to enhance information seeking, targeting and integration. Metadata thus is nowadays the fabric of the Web. Understanding the different forms and arrangements of metadata is in consequence a required skill for researchers and practitioners that aim at understanding and getting value from the Web.

This chapter provides an overview of metadata typologies and uses, clarifying why sometimes different approaches to metadata types and metadata uses could be misleading. We present a general overview of how metadata can be interpreted differently, and then elaborate specific typologies of metadata schemas, metadata schemes, and metadata elements. The second part of the chapter focuses on how these metadata schemas, schemes and elements, which are in a sense abstract constructions and containers, are being deployed by their user communities. Concluding, the analysis of concrete metadata uses discussed within the second part provides new refinements to the metadata types and vocabularies discussed in the first part.

The value and cost of metadata are highly dependent on the functions enabled by metadata inside systems and organizations. As the number of functions and the kinds of organizations that typically create and use metadata are extremely diverse, valuing and cost-justifying metadata turn out to be a difficult and multi-faceted task which involves several different aspects, such as, for instance, metadata language complexity, system target user population, metadata source and metadata quality. This chapter relates these aspects showing some typical cases of organizations using metadata, and then explores how metadata as an organizational asset can be framed in existing Information Systems theories.

The level of service that metadata-based systems could provide is directly influenced by the metadata characteristics. Several of these characteristics are encompassed under the concept of “quality”. As important as it is for these systems, metadata quality is a poorly researched area due to its inherent complexness and subjectiveness. This section presents the main theoretical approaches to the definition of metadata quality and its multiple dimensions. It also presents and compares the most followed frameworks to estimate the quality of metadata records. To provide a real perception of the current status of metadata quality, this section also describe several manual and automated studies in real digital repositories. Full examples of how these studies are conducted are also provided. Finally, the section presents a selection of strategies and tools to evaluate the quality of metadata. The section closes with conclusions about the current state of research on metadata quality and perspectives for the future.

The aim of this chapter is to apply Systems Theory – Systems Science – to Ontologies. The subject matter is given a modern context. The chapter contains a brief description of Systems Theory, which today has become what is known as Systems Science. Subsequently, Systems Science and systems are defined, as well as the links that are responsible for the correct function of a system, known as vectors. Ontologies will be discussed, along with their various definitions and characteristics. This chapter will be based on the premise that Ontology is a complex, conceptual, empirical classification system. An Ontology is structured like a system in which the principal and primary node is the word. A summary will be made of how this system should be constructed, concentrating on the vectors that help the system function. This will be illustrated with an example.

Extensible Markup Language (XML) is a meta-language for defining new languages. Its impact on the modern and emerging web technologies has been (and will be) incredible and it has represented the foundation of a multitude of applications. This chapter is devoted to the presentation of XML and its applications. It provides an introduction to this wide topic, covering the principal arguments and providing references and examples.

In the first part of this chapter, we give a general overview of the most relevant existing approaches to modeling metadata, namely, controlled vocabularies, taxonomies, thesaurus and ontologies. Later the discussion focuses around ontologies, paying careful attention to explaining various key aspects such as the relation between ontologies and knowledge bases, the difference between light and heavy weight ontologies, and the main ontological commitments. Finally, the most relevant semantic web languages currently available are presented. Doing so, a thorough review of RDF, OWL and WSML is provided, where their main variants, underlying formalisms and main applicability are carefully reviewed.

Topic Maps, and their XTM specification, have been revealed as a powerful tool especially designed to describe and organize digital information resources in the last decade. The availability of the ISO standard, and its specialization and revision, reflect the importance of the description scheme, as well as their gathering market and the increasing number of users. The standard specifications have been widely discussed in the specialized bibliography, but a deeper study about the available software collection tools that allow Topic Maps implementation is needed. This work, starting from the Topic Maps definition, details the current available management tools, and comments the integration of Topic Maps with other metadata schemes.

This chapter is meant as an overview of the current state of the art in the area of knowledge engineering, with a particular focus on the methodologies describing how the process of creating ontologies and, more general, semantic content, should be performed in a rigorous and controlled manner. We introduce the most important methodologies proposed in the past years in this field, and the tool environments supporting their application. As a result of this survey we identify open issues and possible directions of research and development which are likely to lead to a mainstream adoption of the methodologies, and thus encourage the industrial takeup of semantic technologies.

This work aims to provide an overview of knowledge representation in e-Learning and concretely, the use of metadata and ontologies to express semantics in educational content. The chapter begins by describing relevant metadata standards for interoperability and reuse of educational resources Developments are also presented in relation to the development of ontologies to improve the use of resources in e-Learning. Emphasis is placed on the methods, representation languages, tools and types of ontologies developed or in use. Thus, the chapter provides guidelines for a semantic approach in E-Learning.

The availability of information technology solutions for data storage and elaboration offers great opportunities to clinical medicine, but requires mastering the complexity of biomedical data.

Biomedical ontologies are often perceived just as a tool for terminological standardization useful for the exploitation of the electronic medium (for queries, statistical analysis, etc.). However, the topics posed by the electronic health records development to the bio-ontologies design also cover more conceptual and epistemological matters; primarily, the need to offer to the clinical user a view of the clinical data compatible with his mental model and work organization. For this reason a more articulated perspective must be adopted to highlight the foundational concepts and perspectives involved into the bio-ontology commitments. Such a critical inquiry over the roles and scopes of ontologies and metadata modeling for health, can be then profitability applied to the description and analysis of the existing general purpose medical ontologies, disease classifications, terminologies and data transmission standards.

The agricultural domain contains multiple different topics that are extensively researched, thus producing a great amount of data and information. The need of efficient organization of this material favors the use of Knowledge Organization Systems (KOSs). The present study carries out an online survey of KOSs that are being used in the agricultural domain so that a representative sample is collected. Then, an analysis of their characteristics is conducted and the main findings regarding the identified KOSs are presented, aiming to provide an initial insight to existing agricultural KOSs and their intended applications.

The post-genomic era is producing an enormous volume of data. Efficient applications and protocols are necessary in Bioinformatics to deal with this information. One of the most-challenging goals is the integration of knowledge extracted from different biological databases. Emerging ontologies in Molecular biology are very promising tools to provide sequence annotation standards that can be shared among genome annotation projects. The Gene Ontology (GO) is the most popular vocabulary to assign biological functions to genes, accordingly to evidence obtained from literature or computationally inferred. GO characterization of gene products is now an essential step of each genome annotation pipeline. Genome curators can use, in addition, auxiliar ontologies to describe biological sequences and alignments. The Open Biomedical Ontologies (OBO) consortium, a joint effort of bioinformatics and biomedical communities, has recently defined a common framework to standardize the ontologies developed in different fields. This chapter provides a comprehensive description of GO and other similar ontologies to annotate genome products.

In this chapter a conceptual and computational framework for the development of Knowledge–Based Engineering tools is presented. The framework is based on the adoption of ontologies and metadata as effective methods to represent the different levels of competence owned by expert designers of complex mechanical objects. After a brief review of literature on this topic, the Chapter gives a complete overview of the approach, explaining how it can be used to represent functional, procedural and experiential knowledge necessary to accomplish a mechanical object design task. Moreover, the Chapter describes a framework application case study, namely the IDS project, to support expert designers working for Fontana Pietro SpA, an Italian enterprise leader in the development of dies for automotive industry.

During last five years we have given a jump forward in the field of emergency management and therefore in the way of focusing the solution. At present, we are including ontologies and metadata for developing this kind of systems, and also there are big efforts to standardize the way that emergencies are communicated and coordinated.

This chapter describes the state of art and efforts to standardize this field.

This chapter focuses on recent research on metadata models and ontologies for tourism. In particular, it presents various ontologies for the tourism domain and research efforts related to semantic metadata models applied to the tourism industry. In this chapter we describe ontologies for tourist destinations, hotels, restaurants, museums, airlines etc. and their related research projects. We survey the current state of semantic web technologies research applied to the tourism domain and identify significant emerging trends and techniques.

This chapter provides an overview of the current state of the art in the area of metadata for multimedia content, focusing on existing metadata standards, vocabularies and ontologies for the description of multimedia content. We introduce the most important metadata standards for images, video, audio and 3D content and provide an overview of recent proposals for vocabularies and ontologies for the semantic description of multimedia content. We introduce typical scenarios in which they are used and outline their actual uses. We conclude the survey with an analysis of open issues and possible directions for future research.

Libraries and information services face the need of dealing with a growing number of metadata schemas. The use of metadata in libraries is not only restricted to resource description; other key activities like resource preservation and the maintenance of complex digital objects also requires the use of additional metadata schemas. In addition, we need to consider the availability of information gateways and databases using heterogeneous metadata. This chapter provides an overview of the different aspects involved in metadata integration and interoperability, from technical aspects related to the capability of exchanging metadata records across heterogeneous systems, to semantic-related issues related to the capability of establishing equivalences between the metadata managed by the different applications. Relevant projects executed in the last years (CORES, Schemas, etc.) and the role of standards like RDF, RDF Schemas or SPARQL in metadata interoperability projects is also described.

The chapter examines two major themes in the area of metadata extraction — formal metadata extraction and subject metadata extraction. The focus of the chapter is on textual documents, with references provided to the multimedia. Approaches to each of the two major themes are presented and discussed, with examples from the area of document classification and metadata extraction from preprints; however, analogous methods exist for use on other types of digital object, and are referenced where relevant. Suggestions are provided in which circumstances and context of use which approaches are most suitable. The chapter concludes with the thought that, owing to recognized evaluation issues, evaluation should be a major future research question in this area.

The subject of metadata and ontology storage is broad, complex and rapidly evolving. This chapter aims to provide an overview of the kinds of technologies that are used for metadata and ontology storage as well as the methods, tools and standards for implementing those technologies. We also include a comparison matrix of metadata and ontology storage and management tools as well as a list of references for locating more information about metadata and ontology storage.