Semantic Intelligence

Here’s some quick thoughts from Centric’s Business Intelligence Practice Lead Jeff Kanel.  Exciting stuff.

For those unacquainted with the term “Semantic Intelligence”, it is the science of relating information through ontologies (conceptual constructs) as opposed to data models.  Mapping ontologies will enable non-technical users (yes, I mean business analysts) to participate in the act of data integration.  New technologies (e.g. RDF triple stores) and semantic data mining enable non-IT people to explore inference-based relationships that could never ever be modeled. Examples are fraud detection, homeland security, social network analysis.

My belief is that business intelligence will evolve into semantic intelligence as vendors can prove value to industry.    Microsoft has led the charge with their BI Semantic Model (BISM) which will be included in the next version of SQL Server (Denali).  Many niche vendors have viable products on the market.  The major barriers are: a) figuring out how to leverage these tools to glean insight, and b) getting industry to take the leap to purchase and build semantic systems.

The impact to us technophiles is that the art of data modeling and data architecture (in particular BI architecture) will become less in demand as the prevalence of semantic systems increases.  Businesses should recognize this is as a potential huge source of competitive advantage and be ready for this sea-change (even though it may take 10 years to fully mature).

Here is a great discussion on emerging issues with data and why semantic intelligence will one day rule.

Here is a better description of what semantic intelligence is.  Better yet just Google it.

I welcome your comments,

Mike Brannan