We’ve heard the quote: Work smarter not harder. Without a doubt, this is one of the holy grails of search. While basic web searches may give the impression search is easy, no one who needs to find that perfectly on-point info-needle suffers any illusions that getting the right results is simple. As chair of the Enterprise Search Summit, I am lucky enough to see presentations on a variety of tools. One of our ESS sponsors, SophiaSearch, will announce the release of a new search solution tomorrow, which uses a Semiotic-based linguistic model to identify intrinsic terms, phrases and relationships between unstructured content.
SophiaSearch CEO David Patterson, walked me through his company’s patented (15 patents, to be precise) approach to search. It certainly strives to understand not only search terms (that someone looking for information on heart attacks will also be interested in myocardial infarctions), but also to better interpret the meaning of a given document based on the body of work that surrounds it.
Sophia’s Contextual Discovery Engine (CDE) is based on the linguistic model of Semiotics, which is the science behind how humans understand the meaning of information in context. The CDE platform detects relationships and themes in unstructured content to enable organizations to search, extract, deduplicate and eliminate redundancy of content to minimize risk and reduce the cost of retrieving, storing and managing information.
While many “smarter” search engines rely on taxonomy to more specifically “understand” the intent behind a search, Patterson says Sophia wants to unshackle organizations from taxonomies. “We don’t believe it is always in an organization’s best interest to conform to pre-existing taxonomies. With our approach, people discover unusual knowledge, things they’ve never thought of before.” He says, “To us, search is two things: the recovery of information that the user expects to find, but more importantly it is that search should be about discovery of new information for the user.”
Initial markets for the tool will be in “sectors with a lot of knowledge workers,” life sciences in particular, such as the Accelrys Pipleine Pilot. However Patterson says the engine can be applied to a wide range of uses, such as Search Engine Optimization (citing an application partnership with Wordtracker), marketing, and the legal vertical (for analysis of IP).
With Sophia’s semiotic approach, Patterson says users can “discover themes in data, build up a picture of relationships” likening the approach to the way in which hyperlinks function on the web.” He says, “Sophia intuitively creates these connections and optimizes results through contextualization.” The tool goes beyond keyword searching to what he calls “Intertextuality… because the meaning a document has depends on all of the other documents around it.”
More details on the tool will appear in tomorrow’s ECXtra newsletter and of course you can learn more about Sophia and many other search solutions striving to deliver tangible business results by joining us at Enterprise Search Summit November 16-18 in Washington D.C.