Sophia is not an internet search engine, but (currently...) is targetted inside the firewall for unstructured documents.
It takes a different approach to most other engines, by using information entropy (from Shannon's pioneering work).
One consequence is that it can find "interesting" documents even if these do NOT contain any of the search keywords which you specify...
Another is that no ontology, tags, dictionary, thesaurus or any other meta-data are required...
The blog post which Darryl gives the URL to gives some more details, and some screen shots based on the New York Times archives.
We don't yet have enough horsepower to offer the service to everyone (need to upgrade our broadband and servers in Belfast)... But if anyone would like a live demo, we can arrange this...
TEDCred score: 0.00 TEDCred reflects your contribution to the TED community.
A reply on Conversation: When will Google's dominance in search end?
Sophia is not an internet search engine, but (currently...) is targetted inside the firewall for unstructured documents.
It takes a different approach to most other engines, by using information entropy (from Shannon's pioneering work).
One consequence is that it can find "interesting" documents even if these do NOT contain any of the search keywords which you specify...
Another is that no ontology, tags, dictionary, thesaurus or any other meta-data are required...
The blog post which Darryl gives the URL to gives some more details, and some screen shots based on the New York Times archives.
We don't yet have enough horsepower to offer the service to everyone (need to upgrade our broadband and servers in Belfast)... But if anyone would like a live demo, we can arrange this...
best wishes - and thanks
Chris