Rensselaer Polytechnic Institute, a private research university located in Troy, N.Y., now has a Watson system all its own thanks to a grant from IBM. Watson, if you’ll recall, is the system that defeated two Jeopardy! champions (and our own Stacey Higginbotham) in 2011. It uses a wide array of data analysis techniques — including, most notably, natural language processing — to understand questions posed in plain English and then searches its vast database for the best answers.
However, despite the obligatory cracks about a rise of the machines or Watson as our new robot overlord, the system is actually limited in its functionality. At this point, it is designed specifically to answer questions and only has access to the data that been uploaded into it (15TB in the case of the Rensselaer system). It’s an impressive feat of software engineering, but Watson is not yet a self-learning system with access to all the world’s data.
Maybe putting a system in place at Rensselaer, and generally giving universities access to Watson, can change that. According to a press release announcing the Rensselaer grant:
“Rensselaer faculty and students will seek to further sharpen Watson’s reasoning and cognitive abilities, while broadening the volume, types, and sources of data Watson can draw upon to answer questions. Additionally, Rensselaer researchers will look for ways to harness the power of Watson for driving new innovations in finance, information technology, business analytics, and other areas.”
The latter effort — harnessing Watson’s capabilities in business — is an important corollary to the computer science work of expanding Watson’s functionality. Early talk (and implementations) is all about using Watson to tackle low-hanging fruit such as health care and law where there are mountains of research papers, case law, medical histories, etc., that no single human could possibly read and professionals could benefit from having Watson point them in the right direction, or perhaps identify some information of which they weren’t aware.
But clearly, as we’ll highlight at our Structure: Data conference next month in New York, big data is a problem for companies in all fields and affects everything from business intelligence to logistics. IBM, for example, is already experimenting with Watson in the retail world. The idea of tuning the power of Watson to some of those problems should have CIOs drooling.