Posts Tagged ‘pipa’
This morning, Lotusphere 2012 had a very special guest, one whose vision and insight changed the world as we know it, but also my own world, helping create a career path that heretofore didn’t exist.
I’m talking about none other than Sir Tim Berners-Lee, the founder of the World Wide Web.
And how ironic that Sir Berners-Lee was speaking to the Lotusphere faithful about the open, semantic Web on a day when so many are protesting the controversial Stop Online Piracy Act (SOPA, as it’s come to be known) as a means towards protecting intellectual property online.
(If you’re interested in learning more, Google has put up a landing page to explain their perspective on the legislation, but in the meantime, properties ranging from Wikipedia to BoingBoing have gone dark and silent in protest.)
As for Berners-Lee message, it was both history lesson and reminder from the past that’s what past is prologue. After Vinton Cerf invented TCP/IP to create the “internetwork” of all those computers, it was Berners-Lee who figured out a way to link all those computers in a more user-friendly (through the HTTP protocol via the WWW).
Now, we’re moving ever closer to the Semantic Web, where not only people, but machines, can understand instructions so we’re eliminating even more friction and sharing that much more information. Berners-Lee seemed to throw a bit of a dart at the siloing of new unstructured data, like certain social networks who have walled off much of their data, but he seems bullish that the continued need to separate data from applications will differentiate the value of that data.
By way of example, Berners-Lee explained that people should be able to look at the same map, on Google Maps for instance, and the separation of the GPS data from the actual application has been what’s facilitated that.
So things that have previously been in those silos, Berners-Lee suggests, will not enable the same value creation should they stay in those silos, and the new value of social business is having people collaborate with all this information and with one another.
Watson, Come Here!
Next on the stage was Manoj Saxeona, general manager of IBM’s recently created Watson Solutions Software Group.
Hard to believe, it’s been a year since Watson beat the best “Jeopardy!” contestants in the world in a widely televised and celebrated match. Now, Saxeona explained, it was time for Watson to stop playing and get down to work.
As Saxeona explained, currently, businesses are dying of thirst in an ocean of data.
Quick, somebody throw me a POWER7 lifeboat!
What most folks didn’t see behind the scenes during the Watson challenge was all the great technology that made Watson possible.
Watson brings together a set of transformational technologies that cultivate the following:
1. An understanding of natural language and human speech.
2) Generation and evaluation of hypothesis for better outcomes.
3. Adaptation and learning from user selections and responses.
The system is built atop a massively parallel, probabilitistic evidence-based architecture optimized for IBM’s POWER7 processors, so it can process 200 million pieces of information in three seconds, which was the threshold it needed to perform and win at “Jeopardy!”
But what about in your doctor’s office. Could Watson help your physician narrow a wide field of diagnoses into a very specific condition?
Absolutely. In fact, medical information is doubling every 5 years, much of which is unstructured.
So for medical diagnostics, Watson can quickly sift through symptoms presented, along with background information like age and other relative demography, medications the patient is taking, and so forth, and then arrive at a narrower list of possible diagnosis.
It doesn’t replace the doctor. It helps the doctor make a more informed decision.
We’ll just have to wait and see as to Watson’s bedside manners!
Written by turbotodd
January 18, 2012 at 3:40 pm