The Semantic Web
http://www.cs.umd.edu/~golbeck/LBSC690/SemanticWeb.html
http://eprints.soton.ac.uk/262614/1/Semantic_Web_Revisted.pdf
https://en.wikipedia.org/wiki/Semantic_Web
In 1989, Tim Berners-Lee invented the World Wide Web. He designed the structure to mimic human thought processes of free association and associative memory. This structure facilitates directionless exploration. Free association is how we associate seeming distinct concept through a series of related concepts. In lecture, we looked at a possible connection between Nash Equilibrium and NASA. The same train of thought can easily be replicated by clicking between Wikipedia pages. The Web is not a perfect imitation of free association as it is limited by the available hyperlinks on a page. In a more accurate imitation, every concept would link to a different page. In reality, only largest concepts tend to have links. The web was created without a navigation system. Berners-Lee acknowledges this: “detractors pointed out that it could never be a well-organized library; without a central database and tree structure, one would never be sure of finding everything. They were right.” (The Semantic Web, 2001).
In contrast, computers are dependent on strict organization. They require a “well-organized library” and “central database” to function. The disconnect between computers and dis-organization of the Web created a new industry: how best to navigate the web. Google became an economic powerhouse by developing the most effective search engine available.
Tim Berners-Lee had another human-based plan for navigation: The Semantic Web. The underlying design of Semantic Web is to create a computer language that comprehends human language. Each concept is given one, unique label, called URI. A concept can be an physical object (ex. a person) or a relation (ex. “X is a sister of Y”). If a concept has two names, for example “zip code” and “postal code”, the semantic web understands both words as the same unique concept with the same URI. If one word has multiple meanings, it understand each meaning as a separate concept and URI. For example, it can differentiate between “Cook” as name, “cook” as a profession, and “Cook” as the house on west campus. Semantic web is able to parse the URI syntax and comprehended the meaning of a string of URIs.
Currently, Web search is designed around how computers are organized. We must us a kind of “Google-language” to find information. If I want to know the height of the Eiffel Tower, I do not ask “what is the height of the Eiffel Tower?”. Instead I type “Eiffel Tower, height”. In theory, Sematic Web could understand that the numeric value I’m looking for is an object [number] with relation [height of] to another object [Eiffel Tower].
If both of Tim Berners-Lee’s visions could be perfectly implemented, the web would resemble an omniscient human. This person is able is explore stream of consciousness with any topic. If another person wanted to know any piece of information, he could ask the omniscient human in natural, human language.
Semantic Language did not develop the way or at the rate Tim Berners-Lee might have liked. The original paper on Semantic Language was written in 2001. It was revisited in 2006. Today, the concept has spread and is starting to be implemented.