Got a thought or idea for our BI Bloggers ? Email biinactionblog@ebizq.net
BI in Action Blog
|
« Competing on Analytics When There's Too Much Data | Main | A Tussle Over Semantics About Semantics » September 20, 2007Semantics: What's the Deal With That?
How relevant are Semantics to business intelligence and analytics? The irrepressible James Kobileus, a fan of all things Semantic (whom I frequently join in the SOA BriefingsDirect podcasts, aka the Gardner Gang), points to a post by Bill Inmon, the guru of gurus in data warehousing and analytics, who questions the value of semantics to the business: "I admit it. When it comes to Semantics, I don’t just get it. You can call me misguided, an old fuddy duddy, or just plain dumb. In one way or another, perhaps all of those names fit. But at the end of the day, I just don’t understand Semantics." A definition is in order here, but that definition can get pretty fuzzy. Even Wikipedia gets tongue-tied and convoluted trying to explain the concept. Essentially, Semantics is identifying the meaning that is attached to data, whether it's through tagging or associations or other means. It's not that Bill Inmon doesn't understand what Semantics are, but, rather, how they ultimately benefit the business. Folks have been talking up the "Semantic Web" (or Web 3.0) a lot lately, which, in theory, puts a great deal of data analysis capabilities out in the cloud versus having to buy, install, and maintain a lot of analytic tools on your own. Jim noted that one of things he likes about Semantic Web "is that one can pretty much connect it to anything one wishes on a philosophic, ontologic, metaphysic, technologic, or sociologic level." In other words, it throws open vast storehouses of information -- formerly locked away from view -- and enables any type of interpretation or leveraging the end-user sees fit. Isn't that ultimately the vision of BI and analytics in its purest form? To be able to ask any question, of any information source, at any time, and get an answer back. An InformationWeek article from May discusses how a couple of companies are already putting some aspects of Semantics into practice. At GlaxoSmithKline, an abstraction layer of semantic data is being tested in an effort to provide a more flexible IT infrastructure and increased productivity through automation. The goal, the article states, is to apply computer-based reasoning to evaluate and filter massive amounts of experimental data. At Eastman Kodak, Semantic Web tools are helping customers manage growing collections of digital images by inferring meaning from visual data. As Kodak chairman and CEO Antonio Perez put it, the goal is to have pictures "begin to recognize each other--so, without human instruction, a picture will use its metadata to find another picture with related metadata, so that all the pictures keep assembling in new groups, depending on how they relate to each other." Lorraine Lawson of IT Business Edge also puts Semantics in their proper perspective, quoting Uche Ogbuji, partner at Zepheira, a knowledge management company specializing in Semantic Web standards. Ogbuji stated (or understated) that Semantic Web technology “has suffered a lack of pragmatic focus.” However, "where the technology really shines, at least at this stage of development, is when it’s used with enterprise data architecture." Currently, most enterprise data architecture efforts are vendor-driven, rather than standards- and Internet-driven. Still, these are pilot efforts, and Bill Inmon ponders whether Semantics will ever show value in the market. "Perhaps semantics just doesn’t have a commercial application and never will. Perhaps the point of semantics is not to have commercial success. Another possibility is that semantics is a technology whose time has not yet come." Posted by joemckendrick in Business Intelligence | Digg This | Add to del.icio.us Trackback Pings TrackBack URL for this entry: |
