Nstein announced this week the release of a new semantic search product they call Semantic Site Search (or S3 for short). What they are doing is standard entity extraction (proper names, places, organizations and so forth) and combining this with sentiment analysis, which creates a semantic finger print that helps expose underlying nuances and meaning in content.
This is in stark contrast to a typical statistical-style search, which may not produce the most relevant results, just the most popular ones. Nstein CTO Jean-Michel Texier said "3S leverages Nstein's patented text-mining technology to power a faceted site search which returns highly accurate results that are organized categorically."
He added, "This has the effect of offering users very relevant related content, ensuring that they not only find what they're looking for but also discover other valuable content.”
Interestingly enough, Nstein's first S3 customer is not a typical enterprise setting using the tool to find information internally, as you might think, but instead as a way to manage web site content, using the technology to create topical microsites based on search results. They also provide a series of out-of-the-box widgets such as "similar content” and “most recent content" that lets users find the content they need (or you want them to find).
This looks like a very interesting product that could be used in a variety of ways. Nstein sees it as much as a content management tool as a search tool, which means they're reaching a point where the two technologies are combining to make a hybrid tool that combines the functionality we have typcially thought of as belonging to one category or the other.
For more information:
- see the Nstein press release [1]
Links:
[1] http://www.fiercecio.com/press-releases/nstein-technologies-launches-semantic-site-search