When we think of the systems that make up a smarter planet, what typically comes to mind are industries like manufacturing, transportation, energy, or banking. But there is another ‘industry’ that needs to become smarter. We might call it the humanitarian industry. That is, the system that creates a safety net to support society and is made up of philanthropies, social services, education organizations, NGOs and government agencies.
In many ways, this is the most human of all systems. So it is ironic to consider how Watson, a computing system, could help us solve civic, social and cultural challenges and make smarter humanitarian decisions. But Watson’s deep QA technology presents new possibilities to do just that. Through private sector collaboration with nonprofits, Watson can become the next innovation to be used as a force for societal good.
From the IBM Watson Research Team: During Watson’s participation in Jeopardy! last week, we received a large number of questions (especially on reddit!) about Watson, how it was developed and how IBM plans to use it in the future. Below are answers to some of the most popular questions (you can also read the responses on the reddit blog). As background, here’s who’s on the team. Continue Reading »
The excitement about IBM’s computer, Watson, and its appearance on the Jeopardy! game show rose to a feverish pitch as the man versus machine drama played out on television earlier this week. The machine won–by no means a foregone conclusion. The episode proved that a small team of highly-motivated geniuses backed by an ambitious, deep-pocketed corporation can create a machine capable of beating the most expert of humans at a sophisticated mental game. It is truly a remarkable moment in the history of computer science and innovation.
Yet in Stephen Baker’s book about the contest, Final Jeopardy: Man vs. Machine and the Quest to Know Everything, the most interesting questions are not about machines but about humans. This is intentional. On page 18, Baker writes that, whether the computer won or lost, it is his hope that it “might lead millions of spectators to reflect on the nature, and probe the potential, of their own humanity.”
Well, Watson beat the human champions in the first game of the Jeopardy! face off between man and machine, with a score of $35,734 to $10,400 for Brad Rutter and $4,800 for Ken Jennings. But Watson’s developers were puzzled by his flub in the Final Jeopardy! segment. The category was US Cities, and the answer was: “Its largest airport was named for a World War II hero; its second largest, for a World War II battle.” The two human contestants wrote “What is Chicago?” for its O’Hare and Midway, but Watson’s response was a lame “What is Toronto???”
How could the machine have been so wrong? David Ferrucci, the manager of the Watson project at IBM Research, explained during a viewing of the show on Monday morning that several things probably confused Watson. First, the category names on Jeopardy! are tricky. The answers often do not exactly fit the category. Watson, in his training phase, learned that categories only weakly suggest the kind of answer that is expected, and, therefore, the machine downgrades their significance. The way the language was parsed provided an advantage for the humans and a disadvantage for Watson, as well. “What US city” wasn’t in the question. If it had been, Watson would have given US cities much more weight as it searched for the answer. Adding to the confusion for Watson, there are cities named Toronto in the United States and the Toronto in Canada has an American League baseball team. It probably picked up those facts from the written material it has digested. Also, the machine didn’t find much evidence to connect either city’s airport to World War II. (Chicago was a very close second on Watson’s list of possible answers.) So this is just one of those situations that’s a snap for a reasonably knowledgeable human but a true brain teaser for the machine.
From a drama point of view, the first episode featuring IBM’s Watson computer competing on Jeopardy! couldn’t have been better. The producers broke the first game into two pieces so they could give viewers a taste of the Watson back story on day one. So the first episode ended with a Single Jeopardy! score: Watson $5000; Brad Rutter $5000; Ken Jennings $2000. Cliff hanger!
At the beginning of the game, Watson got off to a fast start, crisply answering questions about Beatles songs and literary characters. He got the first Daily Double with his answer of “Hyde”–identifying the evil character in the Robert Louis Stevenson book The Strange Case of Dr Jekyll and Mr Hyde.
The humans came back in the second segment, beating Watson to the buzzer on a couple of questions that the machine knew the answers to but wasn’t quick enough to get a shot at answering. Watson showed some panache with his valiant attempt to pronounce the name of the character Jean Valjean, from Victor Hugo’s Les Miserables, with a French accent.
Those of you who watched the contest: What did you think?
By Stephen Baker
Picture the smart, unassuming person at a meeting, who says, “Well, I’m no expert, but once I saw this case where….” That person is doing something that until recently was uniquely human: Soft-pedaling an idea.
Humans beings can soft-pedal because we know what we know (or at least think we do). We also know what we don’t know. And then there’s this entire domain of knowledge in which we know a thing or two. That gray area in the middle is important, because that’s where we can dabble. We can come up with insights and discuss them with people who are better informed. This process widens a discussion beyond the cloistered world of experts. It can lead to insights, the generation of hypotheses, and innovation.
One of the very special aspects of Watson, IBM’s Jeopardy computer, is that it “knows” what it doesn’t know–or, more precisely, simulates this knowledge through statistical analysis. Looking beyond Jeopardy, to Watson’s career in business, this gauge of its confidence is one of its most valuable features.
Say Watson is working in a hospital emergency room. A person comes in with a combination of symptoms that no one has seen before. Someone lists them for Watson. (The machine doesn’t have voice-recognition now, but that could be engineered in a matter of days.) Watson scours its base of documents and research papers, finds various combinations of these symptoms, and lists possible diseases or disorders that the person might have. Each one is accompanied by a confidence gauge. It turns out in this case Watson has only 14% confidence in, say, lupus.
That 14% amounts to a big shrug of Watson’s electronic shoulders. It does not know and is admitting as much. And maybe the doctors have done tests and know that it’s not lupus. But maybe below Lupus, with only an 8% confidence rank, is some other disease that they hadn’t considered. It may be wrong. It may be idiotic. But it may also lead to a thought, a connection. After all, that 8% came from some combination of the symptoms that Watson found in its research. In effect, Watson–even in its ignorance–has come up with a list of hypotheses (along with pointers to its sources). If even one of these hypotheses nudges doctors toward a correct diagnosis, the machine has provided a service–even without “knowing” the answer.
Editor’s note: The following is a guest post from Dr. David Ferrucci, Principal Investigator, DeepQA/Watson, IBM
Just as IBM set its sights on defeating a chess Grandmaster with Deep Blue in 1997, the company’s scientists have developed a Natural Language Processing, Question Answer machine, named Watson (after company founder Thomas J. Watson, Sr.), to challenge two of the world’s trivia grand masters, to be aired on U.S. television from February 14-16, 2011.
Win or lose on national television, Watson will answer the immediate questions, “does it answer questions accurately?” and “does it answer questions quickly?” with a resounding “yes.”
Beyond excitement for the match itself, the team of IBM scientists is motivated by the possibilities that Watson’s breakthrough computing capabilities hold for building a smarter planet and helping people in their business tasks and personal lives. Watson’s ability to understand the meaning and context of human language, and rapidly process information to find precise answers to complex questions, holds enormous potential to transform how computers help people accomplish tasks in business and their personal lives.
Watson will enable people to rapidly find specific answers to complex questions. The technology could be applied in areas such as healthcare, for accurately diagnosing patients, to improve online self-service help desks, to provide tourists and citizens with specific information regarding cities, prompt customer support via phone, and much more.
Like Deep Blue, Watson represents a major leap in the capacity of information technology systems to identify patterns, gain critical insight and enhance decision-making despite daunting complexity. But while Deep Blue was an amazing achievement in the application of compute power to a computationally well-defined and well-bounded game, Watson faces a challenge that is open-ended and defies the well-bounded mathematical formulation of a game like Chess. Watson has to operate in the near limitless, ambiguous and highly contextual domain of human language and knowledge.
Watson’s technology furthers IBM’s leadership in analytics solutions, which help organizations use the vast amount of information they collect to improve their business operations and service to their customers. Additionally, Watson harnesses IBM’s commercial POWER7 system, showcasing how IBM workload-optimized systems provide unmatched capabilities for processing thousands of simultaneous tasks at rapid speeds, once the realm of only scientific supercomputers.
Read more about the technology behind Watson at ibmwatson.com.