Ore.--IBM is installing a more general-purpose version of its
artificially intelligent Watson cluster supercomputer at the Rensselaer
Polytechnic Institute (RPI). The system was specifically architected for
researchers whose aim is to deepen the AI's cognitive abilities as well
as to explore novel new application areas.
original Watson debuted two years ago when it beat the world champions
in the game show Jeopardy. Since then, it has been cloned in versions
designed to give medical advice to doctors, financial advice to
securities traders, and retail advice to businesses.
understands spoken language and can answer ad hoc questions about
knowledge in its domain. Based on IBM's DeepQA technology--which RPI and
seven universities help IBM develop as an open architecture--Watson can
quickly sift through vast amounts of unstructured "Big Data"
drilling-down to the fundamental answers to the queries put to it by
instance, in the healthcare field, Watson can analyze a patient's
medical history and symptoms and in minutes compare them to case studies
in textbooks, hospital databases, medical journals and all the latest
news sources, helping doctors make the most accurate diagnoses and
advise patients about the very latest therapy options available.
Likewise, financial institutions and retailing are applying Watson
similarly to their own Big Data.
will aim to deepen Watson's cognitive abilities by broadening the
number, types and sources of data it can draw on for answers.
Researchers at RPI also plan to explore new applications of Watson in
information technology (IT), telecommunications, business analytics
(BA), commerce and financial services.
Watson will have roughly the same mass storage as the original--15
terabytes--but will allow up to 20 RPI researchers to access it
simultaneously from various locations on its campus. The system was
funded by IBM's Shared University Research (SUR) Award program, under
which IBM will provide Rensselaer with Watson hardware, software and
training. Watson will be located at the Computational Center for
Nanotechnology Innovations, which IBM was a partner in creating at RPI.