September 16th, 2008
Cognition Technologies unveils their Semantic Map
Los Angeles-based Cognition Technologies today announced availability of their Semantic Map to the English language.
According to Cognition’s press release,
“[their] Semantic Map provides software applications with an ‘understanding’ of more than four million semantic contexts”
The press release continues, somewhat opaquely, to expand those ’semantic contexts’;
“It encompasses over 536,000 word senses (word and phrase meanings); 75,000 concept classes (or synonym classes of word meanings); 7,500 nodes in the technology’s ontology or classification
scheme; and 506,000 word stems (roots of words) for the English language.”
Cognition CEO, Scott Jarus, commented;
“Cognition’s extensive Semantic Map is a critical component for the next phase of the Web’s evolution, a.k.a. the Semantic Web, or Web 3.0. It gives the computer a depth of knowledge and understanding of language far beyond the current keyword and pattern-matching technologies in place.”
Moving beyond some of the stand-alone web applications shown on Cognition’s web site, a series of APIs are available in order to permit integration of the Semantic Map into third party Web and enterprise applications.
One of these examples uses Wikipedia and suggests asking questions such as ‘Who caught a cold with fatal consequences?‘
As you might expect from a query selected by the company, the answers are reasonably on-target… and better than those returned by Powerset for exactly the same question.
A ‘definitions box’ clearly displays the way in which the software has interpreted each key word (’cold’ as in the illness rather than as a statement of temperature or emotion, for example), and allows the user to tweak these.
Cognition’s press release points to the resources invested in producing their map;
“These resources were created, codified and reviewed by lexicographers and linguists over a span of 24 years.”
With significant investment also pouring into different-yet-similar efforts at Powerset, Hakia (which has a PubMed demonstrator very similar to Cognition’s health offering) and others, and an Open Source alternative to parts of the Map in OpenCyc, this is clearly one area of the market that still has some settling to do. Each solution is undeniably different, and each clearly has strengths. None, though, is anywhere near perfect, and any one of them may well be good enough for those sufficiently motivated to implement at this early stage. Which will survive, and which will fall by the wayside?
Paul Miller provides consultancy and analysis services at the interface between the worlds of Cloud Computing and the Semantic Web. See his full profile and disclosure of his industry affiliations.
Subscribe to The Semantic Web via Email alerts or RSS.
