Semantifi developed a meaning based search platform to search both structured and unstructured content and filed multiple patents. Semantifi addressed key challenges to searching structured data across the 'search stack':
Semantic search engine driven by knowledge base for relevant answers Automatic data summaries & visualizations census to interpret database results Index multi-terabyte databases for fast results Crawl any type of database to compose complex queries Semantifi's approach is to search the structured web by web hosting its search platform for anyone to use for free and building an App Store like marketplace for a community of publishers to build data search apps for all users to search.
The concept of a "Marketplace for Apps" is clearly proven with App Store, AppExchange, Facebook Apps and so on. Semantifi built SEMANTIFI.COM destined to be such a marketplace for Search Apps At SEMANTIFI.COM, Semantifi's search platform will be centrally hosted for anyone to use for free. Publishers can build their own apps to search any data sets or entire vertical search portals.
Publishers can be individual users, businesses, and content owners. Apps can be of personal interest, or of high public appeal, or have revenue potential. Apps can be shared publicly, or within social networks, or only to paid subscribers. When users search, they get search results from matching Apps. Like at App Store, commercial Apps will generate revenues.
To illustrate, consider an app to research equities. A publisher sources finance data feeds, configures to the search platform and publishes it. Thompson Reuters or Zacks can be such a publisher. A travel portal like Expedia can configure its reservations databases so travelers can see best itineraries, a retailer like J&R could configure its product catalog so shoppers can see products that match its catalog and so on.
A few more ideas of Apps can be an eBay Sellers App to research bid data, a US schools App to research schools performance, a FDIC App to investigate soundness of Banks, an App to search Craigslist Auto/Jobs/Real Estate, etc.
Over 90% of the web remains undiscovered. This is the so-called Deep Web. University of California, Berkeley estimates the Deep Web to be over 500 times bigger than the current surface web. Deep Web goes beyond the billions of web pages and stored inside databases it is largely invisible to current search engines. If only the search engines knew how to search structured data, answers will be readily available to
questions like “Dept of Defense spending between 2000 and 2008” or "Best fare from New York to London next Thursday?". Users like to ask simple questions or keywords to search multiple databases, get relevant answers, see automatic data visualizations in real-time even from multi-terabyte databases. This is the idea behind
SEMANTIFI and the opportunity is ‘orders of magnitude larger than current search’.