June 29, 2007 5:56 AM
Perspective, abstraction and assimilation in web 3.0
Web 3.0 also popularly known as "Semantic Web" is under construction. The current web (web 2.0) is built by connecting documents with embedded links and listed IRIs. These hyperlinks provide a path for a search engine crawler to find other documents. Thus these embedded links and listed IRIs are the main source of information for a search engine that may then use keyword/tags for page ranking. Building web 3.0 web site is different, it requires that published information must be structured in order to add semantics to the information provided. This semantics is added by augmenting information with metadata tags that are publicly used and understood on the semantic web. This public metadata (normative metadata) will relate the different information sources on the web. RDFa provides a mechanism to include RDF vocabulary metadata tags in HTML for publication. This enables information lookup by web 3.0 applications that index on metadata term tags.
What is the advantage of adding metadata term tags to information?
Adding metadata term tags to the information provides semantics and perspective to the information. This perspective provides information to the requester for whom it is meant to be. Instead of being found as one of the lower order web page in million order web pages returned by a search engine for an irrelevant query, the web page will be found for an exact match of the query. The inclusion of normative metadata in the information will allow assimilation of information into the semantic web. By doing so more accurate results can be provided to the user query. Open market is one of the applications that can be built for semantic web, more examples can be found with "text mining" tag. Proprietary XML tags or RDF/OWL vocabularies will inhibit semantic web service providers from including these information sources. Proprietary vocabularies lead to redundant metadata and may require complex service mediators. Normative metadata will allow web services to integrate through service mediators without any human intervention by provider or requester.
|
Example:
A web page that describes a DSL broadband service features may not include the normative metadata terms defined by the DSL forum to describe the DSL commercial services. Thus if there is a popular article on DSL broadband service it need not be found by a web 3.0 application for query "2Mbps DSL subscription", even if this term is used in the article. However the article IRI may be found if metadata tags are embedded in the text. |
How to build a web 3.0 information source?
Define a perspective in order to find semantics to be added to the application. I.e. instead of just providing some information, identify the purpose for providing this information. Once perspective has been defined a vocabulary for normative metadata can be identified. Then apply abstraction resolution to identify the terms that can be translated to metadata tags. Now provide a mechanism to assimilate this information source into the semantic web:
- if this is a service that interacts with other services then provide the machine-readable communication interface definition through Web Services Definition (WSD) and Functional Definition (FD) documents,
- if required submit the web service to a registry or other web application
Assimilation is the amalgamation and modification of newly perceived information and experiences into the existing cognitive structure (AHSMD).
Abstraction is the selection of a certain aspect of a concept from the whole (AHSMD).
Perspective is the appearance of things relative to one another as determined by their distance from the viewer (WordNet® 3.0.).
References:
WordNet® 3.0. from Dictionary.com
The American Heritage® Stedman's Medical Dictionary. from Dictionary.com
I am going to be blogging live from a couple of days of the 
Leave a comment