IBM Watson, the distributed natural language processing platform, isn’t the only advanced system available, but it’s the highest-profile and arguably the most sophisticated. It’s also important to recognize how shrewdly Watson is being marketed. Even before Watson was Watson, IBM was adept at generating publicity for its futuristic computing activities. Most famously, Big Blue beat world chess champion Garry Kasparov in 1997. Four years later, Watson, in a form far more rudimentary than today’s commercialized version, beat Ken Jennings, the human Jeopardy! champ.
Now, Watson is a thriving business and research tool. The marketing continues, with luminaries such as Bob Dylan, Stephen King and Serena Williams helping with television ads. The strategy of equally aggressive marketing and service rollout continues.
Olli, a self-driving 12-passenger shuttle, is being introduced in a Washington, D.C. suburb. Riders will be able to simply tell the shuttle, a product of IBM Watson Internet of Things and Local Motors, where they want to go. It is designed for densely populated areas such as college and corporate campuses. Miami-Dade County has ordered two for a pilot, according to Computerworld.
IBM Watson works to deal with the roads and with riders:
Olli is the first vehicle to use cloud-based cognitive computing from IBM Watson Internet of Things to analyze and learn from 30 sensors embedded in the vehicle. Four Watson developer APIs were used that allow Olli to interact with passengers: speech to text, natural language classifier, entity extraction and text to speech.
IBM Watson marketers tend to paint the platform in mystical terms, as if it can automatically solve any problem. The reality is a bit more nuanced. TechRepublic points to a platform that needs to be trained for as long as 18 months to be optimized. The format with which information is fed into the system is important. Institutions must learn how to interact with IBM Watson and similar platforms. And results are controlled by the source of the data. Thus, if input only comes from specialty institutions – such as The University of Texas’ MD Anderson Cancer Center and Memorial Sloan Kettering Cancer Center in New York City, for example — the resulting information will be narrowly confined to those types of cancers and only cases unique or serious enough to reach being treated at the two facilities.
The idea that the result will be black and white, or yes or no, is false. Suggestions will be offered, and, since the input is human, the suggestions can be incorrect.
IBM Watson is also involved in the fight against diabetes. Last week, the American Diabetes Association and IBM Watson Health announced a partnership at the American Diabetes Association’s 76th Scientific Sessions conference in New Orleans. IBM also used the conference to update its work with Medtronics, according to InformationWeek:
Together with IBM the company is in the final stages of creating a cognitive app called SugarWise. The first generation of the app creates a retrospective analysis of a patient’s insulin, continuous glucose monitors (which are wearables that track glucose levels 24 hours a day), and nutritional data to uncover patterns and trends, according to a prepared statement. The goal is to help people understand how their behavior affects glucose levels in real-time.
IBM Watson tends to generate a lot of news, since it’s a glitzy topic, as far as computing and IT go. And, luckily, it seems there is never a shortage of initiatives, in medicine and otherwise, into which the platform is expanding.
Carl Weinschenk covers telecom for IT Business Edge. He writes about wireless technology, disaster recovery/business continuity, cellular services, the Internet of Things, machine-to-machine communications and other emerging technologies and platforms. He also covers net neutrality and related regulatory issues. Weinschenk has written about the phone companies, cable operators and related companies for decades and is senior editor of Broadband Technology Report. He can be reached at cweinsch@optonline.net and via twitter at @DailyMusicBrk.