Enterprise search is seeing a lot of activity with Google partnering with Dell to manufacture Google's Search Appliance and release of FactSpotter, a search engine from Xerox that uses semantics or "meanings" to do relevant search. However, a clear winner is still elusive.
Google announced that it's turning over the manufacturing of its Search Applicance to Dell (Information Week). The move bores well for enterprise search, as this means the market is alive and kicking. But, are we seeing a dominant force in this market? ... not yet.
In this is very detailed analysis of Enterprise Search products (eight of them) at NetworkWorld, the authors conclude that there is no clear market differentiator. Even Google, whose famous PageRank algorithm made it the defacto choice for Web search, has not yet been able to provide the panacea for enterprise search.
The fact is simple. Enterprises have data in a plethora of formats, where the factors that determine relevancy are dependent on more than subjective data that users or links can provide. Put simply, it's about mapping the meaning in the search queries to the data.
Xerox's recently announced "FactSpotter" (InfoWorld | IDG News Service) is an approach that extracts meaning from the underlying grammar in a query, and the tool can be used across various languages too.
The analysis allows the software to understand that references to "Bill Gates," "he" and "the head of Microsoft" in the same document likely refer to the same person. But the software should also be able to tell that "Bill Gates said ... " and "A friend of Bill Gates said ..." do not precede words spoken by the same person, a situation that would likely lead search engines using keyword analysis alone to return irrelevant results says Frédérique Segond, who manages the parsing and semantics research group at Xerox Research Center Europe.
Semantic search concept itself is not new. Hakia.com is one example of a Web search engine that uses natural language processing to return relevant search results. The bigger question is, who will come up with the right formula to claim the Enterprise Search Crown? What do you feel are the important details to factor for enterprise search? Join the discussion.