Integrating Content Management and Semantics
Background: The aim of the Semantic Web is to encode the meaning and context of key words (concepts) in ways that can be recognised and used by Web applications. Web Agents with a limited and formalised logical capability can discover relationships between concepts, whose meaning and context are defined in dictionaries (ontologies), and thus build correlations between data used originally for different purposes in different sites. Correlations deliver not only faster and more productive searches of published material, but also, and more intriguingly, the opportunity to analyse trends in time and place, with applications including, for example, targeted marketing, law enforcement and data analysis for policy making.
CMS and the Semantic Web: The structure of web sites is not prominent in current discussions about the semantic web. This structure, consisting primarily of page titles, keywords and links, provides a set of additional metadata that can add significantly to the metadata specified by in-text encoding. Content-Management Systems (CMS) add to this four important dimensions:
- Document structure (distinct from web site structure), which forces content into predefined classes of meaning which can be linked to existing ontologies.
- Document history, providing a time frame for the analysis of trends and correlations.
- A much wider authorship than in traditional Web authoring technologies, enabling the creation of user-defined ontologies.
- An easier way of formatting and indexing metadata submitted by users.
Marco Federighi
Faculty of Engineering Sciences, University College London
Torrington Place, London WC1E 7JE, England