Designed some years ago to enrich the search features and the optimal use of results on the web, the technologies behind the Semantic Web are constanly being improved: the progressive integration of these mechanism to search engines and social networks, the development of search engines that meets the specific needs of businesses and professional, a growing use by e-commerce sites: whether to optimize access to relevant information, to improve the ranking of a site or to optimize the marketing strategies, semantic web overlaps with the goals/benefits of the key players of the web and software vendors.
Here's how is a brief overview of its application by year 2011:
The Fundamentals of the semantic web
The semantic web,by extension, the web's metadata, is a concept appeared in 1994, based on a set of technologies intended to make the contents of online resources (pictures, videos, etc.) more accessible and usable by users. Implemented (over 10 years_ by the World Wide Web Consortium (W3C) - the body that standardized web languages-this system can link the information that previously were handled separately. Semantic web is thus inherent to the ability to aggregate multiple data (related to each other) on the web: either via "semantic" or by the attributes that determine them.
For example, the following query: "20 years old new york actor": a metadata engine will be capable of producing a list of items that exactly matches the search criteria requested (name and a set of raw/unstructured data).
From a technical point of view, the Semantic Web introduces several standards:
- RDF (model for describing any data),
- RDF Schema (creation of vocabularies and sets of descriptive terms),
- OWL (language for creating ontologies using support for logical processing (inference, clustering).
- SPARQL (allows us to obtain information from RDF graphs).
The fields of applications of semantic web
The languages behind the Semantic Web are currently used in various fields of applications:
Social networks, blogs and community platforms (eg Flickr, Facebook)
The semantic web can aggregate multiple data between them (eg for an image: the location, date, name of persons, the author, the date a picture has been taken etc..)
To enhance research opportunities for information and/or connect with other members.
We can take the work of the National Library of France, who recently conducted an experiment to demonstrate the potential use of
Semantic Web technologies. It allows you ro simultaneously search across multiple digital heritage collections (in different languages ). The goal is to create more links between the the differnent catalog's data, making them more visible on the web (see here
Searching for information on Internet/Intranet
Search engines designed for a public use are gradually assimilating the technologies of the Semantic Web.
Tumbup is a new search engine connected to Facebook and relying on activity of members of social network (recommendations about places, products, etc..) to produce more meaningful results. Among the other search engine we can include: Wolfram Alpha, True Knowledge (interpretation of natural language) and Zoom, a search engine that can be integrated to the intranet sites and corporate blogs.
Searching for information in an enterprise
Via the software that collect, analyze and organize large volumes of data (eg Exalead, see below)
Research and development fields
Especially in tech fields (eg aerospace, medical research) through the creation of ontologies (a set of concepts and parameters used for a specialized field) to aggregate data design and productionm located in different information systems.
Commerce/ecommerce: through the "GoodRelations" ontology
The vocabulary used to describe structured products, prices, and information related to the business (eg contact details, shops, geolocation, payment options etc..): This program enables search engines to better use of these essential data.
Sematic Web: some examples
Facebook, Open Graph and new marketing opportunities
A few months ago, Facebook introduced the Open Graph Protocol, a technology based on the semantic web, that allows third-party sites to interact with
the social network by sending and receiving information to Facebook. This new protocol is based on the RDFa syntax and aims to "sustain" the social interactions between the visited websites and the Facebook profile of a user. This feature has already been integrated by many sites such as Pandora music catalog: When a Facebook member clicks on the "I like it" button of an artist or a music track on this site, it is added to the list of his favorite music oon his Facebook profile!
The integration of Open Graph is for the time reserved for websites that host data, along which interaction is easily achieved, especially to share common interest and make recommendations: "movies, sports teams, celebrities, restaurants," on Facebook page dedicated to the developer. Eventually, the Open Graph protocol should allow companies to target their ads more effectively to the "fans" who love their products and thus benefit from a more permanent visibility among Facebook profiles.
Google, a step towards the semantic web
Google recently acquired Metadata, the database management company, in order to improve the results generatedby its search engine. It indexes more than 11 million items: movies, books, television shows, celebrities, places, companies and relies on a metadata system intended to allow users to get more information directly from complex queries.
Semantic Web in the enterprise: search optimization on the internet/intranet
In enterprises, applications that arise from the semantic web are mainly represented through research and data structuring tools.
These include Exalead (and its platform Cloud View Finder) and the Antidot Suite/
Exalead and "Cloud View"
The Exalead + CloudView launched two years ago, is a research platform, allowing you to access to information inside and outside your company: It is supposed to collect, analyze and organize large volumes of structured and unstructured data (eg email, RSS feeds, notes, call center, customers reviews, forums, social networks, blogs, etc..) from information systems,mobile or fixed terminals or web. This solution is based on linguistic and semantic features that can classify and enrich the information automatically. The objective is to enable companies to scale their information to meet the different issues.
Antidot Finder Suite and structured search
Available as SaaS (software as service), the seventh version of Antidot Finder Suite (v7) has been recently launched. It is another example of the use of semantic web to meet the needs for the exchange of structured information. Furthermore information extraction (notably via web connectors), is particularly useful for categorization, indexing and linking a set of data from different sources: text files, Open Office
documents and Microsoft Office RSS and Atom feeds, images, sounds, messages and e-mail boxes and has connectors for the web.