Slashdot Mirror

← Back to Stories (view on slashdot.org)

How IBM Plans To Win Jeopardy!

Posted by Soulskill on from the 011010-for-400-alex dept.

wjousts writes "Technology Review is reporting on IBM's plans to take on Trebek at his own game. The 'Watson' computer system uses natural-language processing techniques to break down questions into their structural components and then search its database for relevant answers. A televised matchup with Trebek is planned for next year. 'David Ferrucci, the IBM computer scientist leading the effort, explains that the system breaks a question into pieces, searches its own databases for "related knowledge," and then finally makes connections to assemble a result. Watson is not designed to search the Web, and IBM's end goal is a system that it can sell to its corporate customers who need to make large quantities of information more accessible.'"

117 of 154 comments (clear)

  1. Dealing with Layered Problems by eldavojohn · · Score: 5, Funny

    I wonder how they plan to do with categories that have implications for all the answers. I've seen categories where words must be so many letters in length or perhaps start with certain things and Alex will interject while reading the category such as "'Cats'--and that means all the words in this category start with 'Cat'." Now, with that in mind, a clue could come in as "They are the popular makers of earth moving equipment." Might prompt Watson to find the most popular makers of earth moving equipment--Who is John Deere? The category of 'Cats' would do nothing for Watson without the aid of Alex's interjection ... thus failing at finding "Who is Caterpillar?" (bonus points if you also thought of "Who is Bobcat?" but that answer doesn't start with Cat).

    As a fairly avid though novice crossword puzzler, my mind explodes with questions. Could Watson discern a four letter word for "Pleasant French city" (Nice)? Or what about a four letter word for "Beefy Laker" (Kobe)?

    Lastly, will Watson have something inane and boring to talk about during the break?

    Alex Trebek: Now, Watson, it says here that you are named after Thomas J. Watson who forbade his employees to drink and even frowned upon it while off the job?
    Watson: That is correct. It is against IBM regulation 4-245 Section 8 to consume alcohol on the premises of any facility.
    Alex Trebek: Fascinating, I'm sure you've never broken that strict regulation, ha ha.
    Watson: Good sir, I am a computer, drinking is not within my capacity.
    Alex Trebek: Um, right. So could you tell us something interesting about yourself?
    Watson: *pauses to search records* During the fabrication of my circuitry, several engineers went months without sleep. Leading one to go insane and killed his wife and kid before taking his own life in a double homicide/suicide case.
    Alex Trebek: How unfortunate. Well, I wish you the best of luck today in Jeopardy.
    Watson: Thank you, my snide game show master.

    --
    My work here is dung.
    1. Re:Dealing with Layered Problems by wjousts · · Score: 2, Interesting

      Presumably they will either have to take into account the clues that come from the category itself (as in your example) or rig the system by avoiding "trick" categories. It's not an easy problem and it'll be very interesting to see what IBM come up with.

      An example from last night, they had a category "Knockouts" in both the first and second round. In the first round, all the answers were hot women (i.e. knockouts!), in the second round all the answers were about boxing. How will Watson deal with this? I don't know.

    2. Re:Dealing with Layered Problems by eldavojohn · · Score: 5, Interesting

      Presumably they will either have to take into account the clues that come from the category itself (as in your example) or rig the system by avoiding "trick" categories. It's not an easy problem and it'll be very interesting to see what IBM come up with.

      An example from last night, they had a category "Knockouts" in both the first and second round. In the first round, all the answers were hot women (i.e. knockouts!), in the second round all the answers were about boxing. How will Watson deal with this? I don't know.

      Yes, there are categories which require the contestant to have an active imagination and it's these categories I wish the article had addressed instead of a vanilla one. And I believe it's these categories that makes Jeopardy fresh and new after decades.

      In retrospect, I should have broke out the conversation into a different post so that this wasn't modded +5 Funny. I'm seriously interested in how IBM plans to address things that require the natural speech recognition of Alex Trebek. Does it take into account other answers in the same category to "catch on" like some contestants obviously do?

      Then there's the folks running Jeopardy who could pick some categories that would wreck Watson and give the humans the creative advantage. I hope they exploit this creative ability humans have and write an entire category in ... oh, say Pig Latin!

      In reality, they stand to have much more to gain if the machine comes close to winning ... as they could make this into an annual competition drawing fans and viewers much like the quest to beat the world chess grand masters.

      --
      My work here is dung.
    3. Re:Dealing with Layered Problems by Ethanol-fueled · · Score: 3, Insightful

      It's possible that the questions for that particular show will be specifically chosen to be more explicit and less ambiguous (avoiding the show's characteristic punny wordplay) to put the machine on a more level playing field, keeping its score closer to those of the contestants', which will make the episode more exciting to watch.

    4. Re:Dealing with Layered Problems by eldavojohn · · Score: 5, Insightful

      It's possible that the questions for that particular show will be specifically chosen to be more explicit and less ambiguous ...

      Yes, clues like "It's the cube root of 474552" would level the playing field.

      Isn't the purpose of this to let Jeopardy be Jeopardy? And see if a computer can compete at what the show is?

      --
      My work here is dung.
    5. Re:Dealing with Layered Problems by gnick · · Score: 4, Interesting

      Without speculating on the specifics of tweaking the AI, my guess is that IBM has tried to think through these things. Having put together a few AI bots myself (purely recreationally - you know, just for kicks), I know that I let them play in the real world for quite a while to work out the kinks before unveiling them to nerdy friends and family to show them off and demonstrate just how much time and sleep I'd wasted. My poker-bot played thousands of games in free online rooms before I told anyone that I was even working on him.

      IBM has probably been feeding Watson DVR'd episodes for a while now so that they could identify (if not fix) the kind of gotchas that you're thinking about.

      --
      He's getting rather old, but he's a good mouse.
    6. Re:Dealing with Layered Problems by Anonymous Coward · · Score: 1, Funny

      78 oh shit what is 78

    7. Re:Dealing with Layered Problems by sexconker · · Score: 1

      Well shit, I thought the 1st round was going to have knockouts (boxing) that occurred during the first round, and in the second round, the category would be about knockouts occurring in the second round.

    8. Re:Dealing with Layered Problems by sexconker · · Score: 2, Interesting

      Speech recognition?
      The machine will be receiving a text file of the question.

      Hell, I bet the thing is always the first to the buzzer too.

    9. Re:Dealing with Layered Problems by weszz · · Score: 3, Interesting

      It may not be... would it hit the buzzer and hope it had time to compute the answer like many people on those shows, or would it wait until it had time to compute, and then ring in only if it has the answer?

    10. Re:Dealing with Layered Problems by sexconker · · Score: 5, Informative

      Assume it's a perfect cube.
      x^3 is 6 digits, so we're looking at numbers from about 50 to 100.

      x^3 = 4XX
      6^3 = 216
      7^3 = 343
      8^3 = 512

      70 < x < 80

      x^3 ends in an 2, so the cube root must end in an 8.
      78.

      Seriously though, square roots are easy peasy.
      Cube roots let you use the awesome property that:

      0 - 0
      1 - 1
      2 - 8
      3 - 7
      4 - 4
      5 - 5
      6 - 6
      7 - 3
      8 - 2

      So you can always figure out the last digit of the cube root of a number VERY easily (no, you don't need to memorize that list).

      Then you use the size of the number to get a range, and then estimate. If you're feeling ballsy, you can go for it. Spend the first few seconds (before people buzz in) and get your range down. Then buzz in and spend a couple seconds estimating, then answer (just say "what is..." right when you buzz in). If someone else buzzes in first, more time for you to think.

      4th powers are just doing the square root twice.

      The list for 5th power roots is neat, too.

      0 - 0
      1 - 1
      2 - 2
      3 - 3
      4 - 4
      5 - 5
      6 - 6
      7 - 7
      8 - 8
      9 - 9
      0 - 0

    11. Re:Dealing with Layered Problems by skelterjohn · · Score: 2, Interesting
      Since you bring up crosswords as an example of this sort of issue, let me point you to http://www.oneacross.com/proverb/

      Its an automated crossworld puzzle solver. How it works (and my advisor led the project, though I don't work on anything remotely similar) is that it has a large number of solver modules that are each good at a certain kind of clue. One might be really good at looking up famous people based on keywords. Another might be good at... I dunno some other type of crossword clue.

      Then each of these modules made lists of possible answers for each clue (subject to length and letter constraints), complete with the confidence they had in various answers.

      A central "merger" then collected the candidate answers for each clue from the different modules, and then did lots of tricky search-like algorithms to find a set of answers that seemed the most cohesive.

      This system, PROVERB, was at least the best computer system for solving crosswords, at one time, and did fairly well in competitions in which other humans competed too.

      With Jeopardy! something similar could approach this issue, as well, except without the added constraint that questions to different answers have to relate to each other on the level of spelling.

    12. Re:Dealing with Layered Problems by PMuse · · Score: 1

      I wonder how they plan to do with . . . categories where words must be so many letters in length or perhaps start with certain things . . .

      If they don't play one or more of these categories, it won't be full-up Jeopardy!(tm).

      --
      "We reject as false the choice between our safety and our ideals." --The American President (20.1.2009)
    13. Re:Dealing with Layered Problems by OrangeTide · · Score: 1

      Seeing as you've thought of this by idly considering the problem, I would be extremely surprised if people who have spent months of research on the problem didn't already cover this one.

      --
      “Common sense is not so common.” — Voltaire
    14. Re:Dealing with Layered Problems by ademus05 · · Score: 1

      These tricks work for more than just base 10. You can prove it with Euler's generalization of Fermat's Little Theorem. http://en.wikipedia.org/wiki/Fermat's_little_theorem

    15. Re:Dealing with Layered Problems by Rary · · Score: 1

      it could buzz before it even has the answer calculated.

      So can players. And they do quite frequently.

      --

      "You cannot simultaneously prevent and prepare for war." -- Albert Einstein

    16. Re:Dealing with Layered Problems by sexconker · · Score: 1

      Seeing as it can read the question instantly, instead of having to wait for Trebeck to read it out (or read it as text at human speed), I think it's got a serious advantage.

      I don't think the computational I/O speed is really going to be a limiting factor in this. The crux will be the dataset it has and whatever kind of strategy (being more or less gutsy with the buzzer, bidding more or less on the daily doubles / final round, etc. based on confidence of being right) it uses.

      It's not like playing chess against the Russians.

    17. Re:Dealing with Layered Problems by the_one(2) · · Score: 1

      Well knockout is a synonym for "hot babe". Shouldn't be too hard (of course I'm sure there will be something else that will be very hard for it)

    18. Re:Dealing with Layered Problems by sexconker · · Score: 1

      Which trick?

      If you're referring to finding out the last digit, it certainly DOESN'T always work out that nicely.

      For squares we've got 3 and 7 both ending in 9. 8 and 2 both ending in 4. 9 and 1 both ending in 1. 4 and 6 both ending in 6.

      Similar shit streams on down the ass for any even-power situation.

      Does this trick hold true for all odd powers? I've got a feeling that it does, and would do so for any base. Couldn't be arsed to try to prove it though.

    19. Re:Dealing with Layered Problems by ThisIsForReal · · Score: 1

      If you buzz at the wrong time, you get locked out for about a second.

      --
      -THE END-
    20. Re:Dealing with Layered Problems by Garrett+Fox · · Score: 1

      I'm on a mailing list with some participants in the Loebner Prize Contest, a version of the Turing Test. ("Robitron," if you want to see the list.) We were talking about this Jeopardy AI project recently.

      The most-discussed type of AI on that list is a "chatterbot" like ALICE or the classic ELIZA -- one that basically looks for key words and finds pre-written responses to them. That approach could probably tackle a variety of Jeopardy-type questions if someone took the trouble to feed the AI a bunch of suitably-formatted stimulus/response pairs. Still, I've argued that that type of AI is a dead end both for understanding the mind and for most practical purposes -- and the IBM approach sounds similar to it.

      IBM is known for its chess computer, Deep Blue/Deep Thought, which used a brute-force method of searching a huge database of possible board positions. Since there are 400 possible board positions in chess after just one round, memorizing all possible layouts is absurdly data-intensive and almost certainly not how humans play chess or do most other things. (The poor progress of AI for the game of Go, in which there are many more possible layouts, also suggests that brute-force lookup isn't how we do things.) I'm more interested in the AI approach described by Douglas Hofstadter in "Fluid Concepts and Creative Analogies," which focuses on lots of semi-random "codelets" doing lots of processing on limited data. Using that style for a chess AI would involve having it make observations like "this is one of those knight-fork situations."

      The upshot is that while IBM might be able to play a passable game of Jeopardy if treated as deaf and mute (given written I/O), that project won't get us much closer to human-like AI.

      --
      Revive the Constitution.
    21. Re:Dealing with Layered Problems by ademus05 · · Score: 1

      Sorry, I was ambiguous. Fermat's little theorem says a^(p-1) mod p = 1 mod p, and a^p mod p = a. When p = 5, this means that any number raised to the 4th power is going to end in a 1 or a 6. Euler's extension to Fermat's little theorem says that a^(t(n)+1) mod n = a mod n, where t(n) is a function. t(10) = 4, so this explains why the last digit of a number is unchanged after raising it to the fifth power. t(100) = 40, so you'll find that after raising a number to the 41st power, the last two digits will be the same.

    22. Re:Dealing with Layered Problems by JustPutt2 · · Score: 1

      If you had recently noticed, the IBM announcement about what they call "Stream Computing" you might have a chance to see something new.

    23. Re:Dealing with Layered Problems by JustPutt2 · · Score: 1

      I will say it again,,, watch out for what IBM calls Stream Computing,,, they may do a blow away like they did in the Chess match.

    24. Re:Dealing with Layered Problems by sexconker · · Score: 1

      Doesn't work for 2 or 5...

      2^41 = 2199023255552; % 100 = 52. 2 % 100 = 2.
      52 != 2.

      5^41 = 45474735088646411895751953125; % 100 = 25. 5 % 100 = 5.
      25 != 5.

  2. A test with Wolfram|Alpha by Anonymous Coward · · Score: 2, Interesting

    I fed all the Jeopardy questions into Wolfram|Alpha and it got every single one right.

  3. Jesus by BadAnalogyGuy · · Score: 2, Funny

    What was an extra-terrestrial?

    1. Re:Jesus by wirelessbuzzers · · Score: 1

      "My kingdom is not of this world" indeed...

      --
      I hereby place the above post in the public domain.
    2. Re:Jesus by treeves · · Score: 2

      I know you were going for funny, but in reality Jeopardy never uses one(or two)-word answers. There are too many possible questions for a one-word answer.

      Example:
      A: Caesar

      Q: Who crossed the Rubicon and said 'the die is cast'?
      Q: Who said 'Et tu Brute?' when he was assassinated?
      Q: Which Roman emperor adopted his great nephew Octavian who later became Augustus Caesar?
      Q: What salad is made with Romaine lettuce, anchovies, garlic and lemon?
      etc. etc.

      Instead you use the long answer and get a short question:

      A: This Roman emperor adopted his great nephew Octavian who later became Augustus Caesar.
      Q: Who is Julius Caesar?

      --
      ...the future crusty old bastards are already drinking the Kool-Aid.
    3. Re:Jesus by mopslik · · Score: 1

      I know you were going for funny, but in reality Jeopardy never uses one(or two)-word answers.

      They often do, but they're typically framed within a narrow, specific category. For example, the category might me "National Drinks" or some such thing. Typical Q/As might be Japan (Sake), Russia (Vodka), and so forth. Jeapoardy! also has a few other categories (anagrams, for example) that frequently use one or two words.

    4. Re:Jesus by treeves · · Score: 1

      Ah, you're right. I forgot about those cases. But they are exceptions, not the norm. I haven't watched Jeopardy! in a long time. I always thought I would have done well on Jeopardy! but not as well as that guy from Utah.

      --
      ...the future crusty old bastards are already drinking the Kool-Aid.
  4. Only if... by weszz · · Score: 5, Funny

    It can answer in Sean Connery's voice and make your mother jokes at him.

    Otherwise I'll probably pass and look up old SNL skits on youtube instead.

    1. Re:Only if... by SterlingSylver · · Score: 5, Funny

      So I think IBM's plans here are to
      Use a high-tech set of
      Computers to create a
      Knowledge processor that can be monetized.

      I think
      That wanting

      To use such a
      Rediculously advanced
      Engineering marvel to make Sean Connery jokes would
      Be a waste of
      Everone's time, energy, and
      Karma

    2. Re:Only if... by weszz · · Score: 1

      think of it this way, if it can make jokes as well as being that good, I believe it would make more money and add some unexpected comic relief to things making it that much more valuable.

      you don't need to put your A people on it, just some guys that know what they are doing may do it on their time as well...

    3. Re:Only if... by Ogive17 · · Score: 2, Funny

      IBM> You'll rue the day you crossed me Trebek!

      --
      "Action without philosophy is a lethal weapon; philosophy without action is worthless."
    4. Re:Only if... by wjousts · · Score: 1

      They can't also have some fun with the snobbiest host along the way.

      I never thought Trebek was snobby at all. Did I miss something?

      Maybe Jeff Foxworthy is more your speed?

  5. How IBM Plans To Win Jeopardy! by Anonymous Coward · · Score: 1, Funny

    They plan to answer "Kebert Xela" and send that bastard back to the dimension where he belongs.

  6. Suck it Trebek! by Anonymous Coward · · Score: 5, Funny

    I wonder how well it'll do at Anal bum cover.

    1. Re:Suck it Trebek! by Volante3192 · · Score: 4, Funny

      Clearly, some mod has never seen SNL.

      (That's 'An Album Cover', Connery!)

    2. Re:Suck it Trebek! by weszz · · Score: 2, Funny

      or the other classics...

      Connery - "I'll take the PEN IS mightier for 300..."
      Trebek - "That's the Pen Is Mightier..."

      Connery - "I'll take Famous Tities for 500"
      Trebek - "That's Famous Titles..."

      It's not so easy when you don't have the answers to look at is it Trebek?

    3. Re:Suck it Trebek! by Anonymous Coward · · Score: 1, Informative

      A: The answer is, what do you call someone who can spell neither penis nor titties.

      Q: What is weszz?

    4. Re:Suck it Trebek! by 0100010001010011 · · Score: 2, Informative

      Thank Goodness for Hulu (instead of finding dead Youtube Links)

      The wiki entry is also pretty good: http://en.wikipedia.org/wiki/Celebrity_Jeopardy!_(Saturday_Night_Live)

      "Catch These Men" for "Catch the Semen".
      "Things Trebek Sucks" over the actual category, "Potpourri".

    5. Re:Suck it Trebek! by bitfarmer · · Score: 1

      I wonder how well it'll do at Anal bum cover.

      I think "Anal bum jacket" is way funnier...

      --
      Eagles may soar, but weasels don't get sucked into jet engines.
  7. I get it now... by click2005 · · Score: 4, Funny

    sell to its corporate customers who need to make large quantities of information more accessible.'"

    They want to replace the call centres in India with call computers.

    "Hello you're speaking to Susan Blue Gene how can I help you?"

    --
    I am a free slashdotter. I will not be modded, blogged, DRM'd, patented, podcasted or RFID'd. My life is my own.
    1. Re:I get it now... by internerdj · · Score: 1

      http://www.despair.com/motivation.html/

    2. Re:I get it now... by iamhigh · · Score: 1

      I think you were looking for this one... http://despair.com/innovation.html

      --
      No comprende? Let me type that a little slower for you...
    3. Re:I get it now... by maxume · · Score: 1

      Well, those reps usually aren't empowered to do anything that the computer won't let them, so as long as the recognition is somewhat consistent, who cares?

      --
      Nerd rage is the funniest rage.
    4. Re:I get it now... by Pie+Pan · · Score: 3, Funny

      I'm suddenly reminded of this: http://www.penny-arcade.com/comic/2000/01/28/

    5. Re:I get it now... by StikyPad · · Score: 2, Funny

      Ah, but how are they to be mimicking the bizarre structure of grammar I am having to become accustomed to?

      --
      https://www.eff.org/https-everywhere
    6. Re:I get it now... by ausekilis · · Score: 1
      More correctly, they'll replace the lists the India call centers read with keyboards to input crap into the computer.

      India: Hello, how may I help you today?

      Customer: My computer has a blue screen

      India: One second while I type that in... (15 minutes later). Have you tried shutting up and restarting your computer?

    7. Re:I get it now... by flyingsquid · · Score: 1
      They want to replace the call centres in India with call computers.

      But if all those computers start malfunctioning, who can we call for tech support?

    8. Re:I get it now... by quanticle · · Score: 1

      The next one is even better: http://www.penny-arcade.com/comic/2000/1/31/

      --
      We all know what to do, but we don't know how to get re-elected once we have done it
    9. Re:I get it now... by perlith · · Score: 1

      Parent got modded "Funny", but should have instead been modded "Insightful" for new business idea to cost cuts. Unless, of course, IBM has already patented that idea...

  8. Jesus by eldavojohn · · Score: 5, Funny

    What was an extra-terrestrial?

    How tastelessly incorrect. Extra-terrestrials don't come back to life. Watson would cross reference The Bible with many recent movies and come up with the correct question we were looking for: "What was a zombie?"

    --
    My work here is dung.
  9. Wordplay by ooutland · · Score: 5, Insightful

    A lot of Jeopardy questions are wordplay-dependent, something AI doesn't have the hang of yet (unless IBM has been toiling in secret on something truly amazing). Categories like "Rhyme Time" and questions like "Qhat does a Pharoah need when he has a cold?" (Answer: an Egyptian Prescription) are beyond the ken of a data search.

    Many Jeopardy "answers" have the key to the answer within the question, though in some cases it may be enough to throw the program off. IE in a category like "Musicals" an answer like "Unlike his other hits, this musical wasn't 'the cat's meow' on Broadway." Raw data crunching will pair musicals, Broadway and "cats" but won't know where to go with "unlike." Only an aficionado will know that Andrew Lloyd Weber's "Starlight Express" tanked on Broadway.

    So the writers, given any knowledge of the limitations of AI, can set a challenge which will be nearly impossible for current AI to meet. John Henry will live another day.

    --
    I'm the queer the atheists sent here to take away your gun!
    1. Re:Wordplay by Another,+completely · · Score: 1

      Right. That's why it's interesting. It needs to interpret the questions, construct a query, process and rank the results, as well as store and index all the information it needs for the game (no live connection to the internet).

      If the questions aren't from the same group as usual, it won't be worth much. I would hope they wouldn't be specifically designed to either help or stump the computer.

    2. Re:Wordplay by PJ1216 · · Score: 1

      It says its not designed to search the web. This doesn't imply that its database must be stored locally. It can have a live internet connection, but only be talking to the database. I'd imagine the database is quite large.

    3. Re:Wordplay by Another,+completely · · Score: 2, Informative

      Sorry, I used information from outside the original article without a citation. This is from the team's web info:

      ... just like human competitors, Watson will not be connected to the Internet or have any other outside assistance.

  10. Waste by Tablizer · · Score: 2, Interesting

    IBM is laying off American citizens, but hiring in Asia, and yet are spending all this money on gimmicks. This is the kind of thing that gives big companies bad names. Hopefully, as a consolation prize, the laid-off Americans can watch their former company go down in smoke on the game show, hoping it starts smoking and sparking like a cheesy Trek android meltdown.

    --
    Table-ized A.I.
    1. Re:Waste by wjousts · · Score: 2, Interesting

      It's not a gimmick. It is very important research on AI and natural-language processing. Jeopardy! just happens to give them a very difficult problem to tackle. If they can develop a system that can handle Jeopardy!, it'll be a huge break through for other fields.

    2. Re:Waste by glwtta · · Score: 4, Informative

      They're spending money on research, gimmicks just help pay for it.

      --
      sic transit gloria mundi
    3. Re:Waste by danger42 · · Score: 1

      I'm sorry, but you didn't answer in the form of a question.

      --
      -nd
    4. Re:Waste by Swanktastic · · Score: 1

      Well, since a big portion of IBM is consulting (50%?), it probably makes sense for them to hire consultants who are citizens of the countries they will be working in... How many Americans are fluent in a second language? How many are willing to relocate for 5+ years to Asia? I'd guess that hiring local gives IBM a GOOD name in those markets.

      Plus, the people they are hiring are human beings too. Why is hiring an American a more noble thing than hiring an Indian or Chinese person? They need to feed their families too... On the converse side, would you take a pay cut to work for an American company instead of a Japanese company out of patriotism? An Amercian who coasted through high school makes more than a Chinese MD who was the best student in his class. Is that fair? The rest of the world sees American workers as being no better than the rich kid who was born into an aristocratic family and lives off his daddy's hard work.

    5. Re:Waste by The+Archon+V2.0 · · Score: 1

      IBM is laying off American citizens, but hiring in Asia, and yet are spending all this money on gimmicks.

      Hardly a gimmick. It's called "advertising". In the public mind:
      (Jeopardy!) = smart
      therefore
      (computer winning Jeopardy!) = smart computer.

      Plus, they get a hype boost just from using the show's name. Would this article have even made Slashdot if it were "IBM plans to have AI accurately answer game show questions"?

    6. Re:Waste by Tablizer · · Score: 1

      Plus, the people they are hiring are human beings too. Why is hiring an American a more noble thing than hiring an Indian or Chinese person?

      Lopsided imbalanced trade creates all kinds of problems, and was a large contributor to the mortgage meltdown. Other countries can increase GDP without relying on the US to be a mass importer of their goods. They just do it out of habit and because we let them. The early US grew even when Europe tariffed us up the wazoo. We mostly traded with ourselves.
           

      --
      Table-ized A.I.
  11. CATS MEANING ALL RESPONSES START WITH CAT by tepples · · Score: 4, Insightful

    Alex will interject while reading the category such as "'Cats'--and that means all the words in this category start with 'Cat'."

    Then the bot would read the closed caption that the category is "CATS MEANING ALL RESPONSES HAVE A WORD THAT STARTS WITH CAT" and include that in its reasoning. Then the clue "They are the popular makers of earth moving equipment" becomes something like "They are the popular makers of earth moving equipment, starting with 'CAT'".

    1. Re:CATS MEANING ALL RESPONSES START WITH CAT by RealGrouchy · · Score: 1

      Hm... one of the categories could be "Cats", you say...

      That must be why IBM didn't want the machine to search the Internet!

      - RG>

      --
      Hey pal, this isn't a pleasantforest, so don't waste my time with pleasantries!
    2. Re:CATS MEANING ALL RESPONSES START WITH CAT by fprintf · · Score: 1

      I loved Cats. I would see it again, and again!

      p.s. For the humor impaired mods, check out the old Saturday Night live clips.

      --
      This post brought to you by your friendly neighborhood MBA.
  12. Bad editors! by sootman · · Score: 4, Funny

    The summary clearly should have been titled "How does IBM plan to win Jeopardy?"

    --
    Dear Slashdot: next time you want to mess with the site, add a rich-text editor for comments.
    1. Re:Bad editors! by doomy · · Score: 1

      They will clearly win Jeopardy by patenting all the words in popular culture and using something like DMCA against all and any opponents... thus enforcing their patent argot.

      --
      ...free your source and the rest would follow...
  13. The end? by LoudMusic · · Score: 1

    A computer that can play Jeopardy?

    THE END IS NEAR!

    --
    No sig for you. YOU GET NO SIG!
    1. Re:The end? by dzfoo · · Score: 1

      No, but you should start worrying as soon as it learns how to play Tic-Tac-Toe with itself. From there on, it's just a quick step into Global Thermonuclear War!

              -dZ.

      --
      Carol vs. Ghost
      ...Can you save Christmas?
    2. Re:The end? by stonefry · · Score: 1

      Would you like to play a game? ... of Jeopardy?

    3. Re:The end? by pacinpm · · Score: 1

      It will be the end... for Jeopardy if IBM wins. Who will want to watch the show where computer is better than humans?

    4. Re:The end? by tod_miller · · Score: 1

      I am glad we can count on your to predict the future.

      > I don't see many sales in the future of iPod.

      http://apple.slashdot.org/comments.pl?sid=22940&cid=2467504

      lol, inb4 20/20 hindsight. Just look at the comments above yours.

      --
      #hostfile 0.0.0.0 primidi.com 0.0.0.0 www.primidi.com 0.0.0.0 radio.weblogs.com
  14. Re:Suck it Trebek by weszz · · Score: 1

    That's S-words for four hundred...

  15. Re:Why is "Watson" such a popular choice of name? by grouse · · Score: 5, Informative

    Because Thomas J. Watson was the man who turned IBM into a global empire, and Thomas J. Watson Jr. brought it into computers. They successively held the top position at IBM for 57 years. So it's a very important name at IBM, and the connection with Sherlock Holmes is serendipitous.

  16. Re:Why is "Watson" such a popular choice of name? by Anonymous Coward · · Score: 1, Informative

    http://en.wikipedia.org/wiki/Thomas_J._Watson

  17. The Computer will lose by flattop100 · · Score: 1

    "I'm sorry, Watson. Your answer must be in the form of a question."

  18. Re:Why is "Watson" such a popular choice of name? by glwtta · · Score: 1

    There are many Watsons. Working in biotech I've seen dozens of machines and applications named Watson.

    --
    sic transit gloria mundi
  19. Infinite Loop by jimmi_hendrix · · Score: 1

    I hope "how many roads a can a man walk down..." is not a question

  20. You'll Rue Day Trebeck!!! by nexxuz · · Score: 1

    IBM just MUST make it sound like Sean Connery! Watson: I Google'ed your mother last night Trebeck!

    --
    I love random hex numbers! Just like this one, 09f911029d74e35bd84156c5635688c0.
  21. I want to see how it responds to this question... by PJ1216 · · Score: 3, Funny

    What is the answer to life, the universe, and everything?

  22. Bullshit by sexconker · · Score: 1

    The system is not designed to access the web?
    Horse shit.
    That huge fucking pile of data is getting in there from the web. It won't be accessing the web during the game, but it's still a fucking cache of random shit (mostly geography and world history) from the internet.

    1. Re:Bullshit by Bigjeff5 · · Score: 1

      So? Your point? It's not accessing the web live, is it? No? So it's not getting it from the web.

      The web isn't some magical information generating device. That information did not originate on the web, somebody put it there either from their own mind or from an offline location. By your own reasoning accessing the web isn't even "getting it from the web", as the web is just a huge cache of information from people's homes, schools, and even just plain their own minds.

      Which is actually, in a sense, correct, and why your whole complaint is nonsense. People store that information in their brains, the computer will store it in a database on a few hard drives. It's no advantage, and the thing has plenty of significant disadvantages in a contest like this, unless they rig the system or IBM has made some serioius frickin advances (which they may have done).

      Quit whining.

      --
      Security is mostly a superstition... Avoiding danger is no safer in the long run than outright exposure. - Helen Keller
    2. Re:Bullshit by sexconker · · Score: 1

      Uh, no.
      A few interns were told to trawl wikipedia to grab data and form it into statements (probably prolog style).

      I guaranfuckingtee it.

      The game IS being rigged for the machine - the answers will be given to it in text. It won't have to press a physical button on a buzzer. Etc.

    3. Re:Bullshit by Bigjeff5 · · Score: 1

      Ok, first, you totally missed my point. My point was that by extrapolating your objections to having a database of information obtained via the web, you actually get the data indirectly from the source, in some form. I don't see your objections to the human players who have been crawling for data for their own databases for the last 40+ years. Some of them have been crawling the web for data for the last 20 years!

      It's the same damn thing, data is data, half of Jeopardy is knowing shit. That's an area computers may shine. The other part is processing the context of the information and plumbing your data for the relevant piece of information in a fraction of a second. That's an area computers really, really suck at. Their data retrieval is slow compared to that of a human, and their ability to understand context sucks monkey balls.

      Having a massive database of data is the only advantage the thing has against a human contestant.

      Lastly, are not the questions being fed to the human players in the form of text as well? In fact the human players get auditory cues as well. I don't understand your objection here. It's not like the machine gets it before hand. Being fed in text form only is a disadvantage, not an advantage.

      The buzzer may or may not prove to be a boon, it could spell disaster if the machine is too agressive. And I don't see the problem of no physical button, as long as it has to buzz in, and have the answer ready in a reasonable time (just like humans). If it can buzz faster but fails to parse the context correctly it is going to lose, and hard. Human players already do this, they buzz early before they have the answer, based on their confidence that they -will- have the answer in the next few seconds.

      Nothing wrong with a machine doing the exact same thing, a physical button doesn't help or hinder that. Hello humans can already buzz before the question is finished if they want, they just had better have the answer.

      --
      Security is mostly a superstition... Avoiding danger is no safer in the long run than outright exposure. - Helen Keller
    4. Re:Bullshit by sexconker · · Score: 1

      No, you missed my point.
      And you said so, yourself:

      "So? Your point?"

      Don't respond to my post, then put your own point in when you don't even get my point, then claim I don't get yours. Seriously, wtf?

  23. Patents by Andrew+Cooper · · Score: 1

    So will IBM then try to get a patent for "Winning Jeopardy", then all the contestants have to pay royalties if they win?

  24. They're doing it wrong by mandark1967 · · Score: 1

    All they need to do is use their super computers to generate some digital footage of Alex Trebek engaged in beastiality (tappin' Rosie O'Donnell) and then tell him that if they win, the footage disappears forever.

    It'd be far cheaper than what they are planning to do...and they can always leak the footage to youtube after they walk off with the winnings...

    --
    Sig Follows: "Suppose you were an idiot. And suppose you were a member of Congress. But I repeat myself." -- Mark Twain
  25. Video Daily Double by Comboman · · Score: 4, Interesting

    hoping it starts smoking and sparking like a cheesy Trek android meltdown.

    Alex: "Here to present the Video Daily Double is Harry Mudd, who always lies."

    Harry: "I am lying."

    --
    Support Right To Repair Legislation.
  26. How IBM Plans To Win Jeopardy..... by Fbelch · · Score: 1

    By searching for all the answer on http://www.wikipedia.org/ because we know all the information on that site is correct!

    (Yes.. that was a joke)

    1. Re:How IBM Plans To Win Jeopardy..... by Bigjeff5 · · Score: 1

      Hey, anybody in the entire world can edit a wikipedia page, so you know you have the absolute best information possible!

      --
      Security is mostly a superstition... Avoiding danger is no safer in the long run than outright exposure. - Helen Keller
  27. Re:I want to see how it responds to this question. by Bigjeff5 · · Score: 2, Insightful

    Well, we already know that, it's 42.

    The real question, is what is the real question for which 42 is the answer? That one is the tough one.

    I suggest we build a planet, who's sole purpose is to calculate that question...

    --
    Security is mostly a superstition... Avoiding danger is no safer in the long run than outright exposure. - Helen Keller
  28. Wrong bot for the job by DeathMagnetic · · Score: 1

    It's a shame that 'Watson' works by breaking down questions into their structural components and then searching its database for relevant answers. After all, on Jeopardy all the answers are given freely.

  29. Comment removed by account_deleted · · Score: 3, Funny

    Comment removed based on user account deletion

  30. It was more exciting by Amazing+Quantum+Man · · Score: 1

    The first time I read it

    --
    Fascism starts when the efficiency of the government becomes more important than the rights of the people.
  31. Please answer in the form of a question... by NotQuiteReal · · Score: 1

    "Hello you're speaking to Susan Blue Gene how can I help you?"

    Making that statement in the form a question was appropriate there... unfortunately all the statements will be in the form of questions, with no answers in sight...

    Is it plugged in?
    Is it turned on?
    Did you reboot?
    Do you have your serial number?
    ...?
    Would you like me to drop this call under the guise of transferring you to someone who has no script to follow?

    --
    This issue is a bit more complicated than you think.
  32. Weird example by RobotWisdom · · Score: 1
    In explaining the chain of reasoning, they weirdly left out that the name 'Pagliacci' is explicit in the lyrics, and proposed that Watson would deduce 'tears' as a form of feelings!? (Maybe they don't want to include a database of song lyrics?)

    They claim they won't use Web data, but there's no way they can compile enough databases on their own to handle Jeopardy's general knowledge. Awards, lyrics, plots, characters... the list goes on and on and on.

    WolframAlpha is a recent disappointment that's spent years collecting databases and delivers almost nothing useful yet.

    I'd suggest celebrity blind-items as a fun test domain that might be manageable, eg:

    "Which female rocker, who was once married to a famous old-school singer, now has a penchant for young girls?" [cite]

  33. How to build a Jeopardy player (the easy way) by professorguy · · Score: 1

    Here's a way to build a simple Jeopardy player that would kick a human's ass and doesn't require 4 years of programming:

    - Type entire "answer" as given on the board directly into google without quotes.
    - Search the returned page for the most common word (ignoring 2 letter ones) in the titles of the pages.
    - If the most common word appears more than 3 times, print "What is X?" where X is the common word.
    - If no one term appears that often, don't ring in.

    Voila. Instant human-crushing Jeopardy player.

    If you tweak the rule set to make it a little more complicated (looking for whole phrases, etc) and tweak the threshhold for how "certain" it must be before ringing in (the appearance count), you might be unbeatable.

    1. Re:How to build a Jeopardy player (the easy way) by Bigjeff5 · · Score: 1

      You would think so, but it's not so simple.

      Google results are far from useful without context information, especially with Jeopardy where the context is usually hidden by a pun or a play on words.

      You just don't notice it because most people are able to process this contextual information on the fly, but it is a huge challenge for a machine to do it. There are litterally thousands of little bits of information that we collect as the answer is being given, including the context of the category, i.e. whether said category was a joke or not, the particular round it takes place in and the general theme of that particular show, if there is one.

      A lot of the information can be given to the machine before hand, but a lot more must be collected and processed on the fly - something human brains do with ease, but computers have yet to conquer that area.

      Contextual stuff is tricky, and we deal with rather complex contextual issues constantly, continually learning from them, making it seem extremely simple to us but is infact some serious mental gymnastics that (so far) machines haven't been able to duplicate. Look at the plethora of useless answers Google often gives when you are looking at that one gem of useful info. Humans can easilly pick out the relevant info, where machines can't (as the google results themselves demonstrate). Maybe the point of the exercise is that the IBM software can? That would be a really neat trick.

      --
      Security is mostly a superstition... Avoiding danger is no safer in the long run than outright exposure. - Helen Keller
    2. Re:How to build a Jeopardy player (the easy way) by OrangeTide · · Score: 1

      Of course, you're solved it, you're so smart. Wait, I mean incredibly arrogant.

      --
      “Common sense is not so common.” — Voltaire
    3. Re:How to build a Jeopardy player (the easy way) by MarkGriz · · Score: 1

      Easier...

      Hide Ken Jennings inside.

      --
      Beauty is in the eye of the beerholder.
  34. Re:I want to see how it responds to this question. by StikyPad · · Score: 1

    Answer: The number of minutes it would take for gravity-powered travel between antipodes, and the angle in degrees which causes a rainbow to appear.

    --
    https://www.eff.org/https-everywhere
  35. IBM and PR by johnwbyrd · · Score: 1

    IBM has a history of inadvertently making terrible PR for themselves with these man-vs-machine stunts. Everyone here should remember Kasparov vs. Deep Blue. Expect IBM to win Jeopardy, and expect there to be a hailstorm of "IBM cheats" controversy after the game.

    http://www.youtube.com/watch?v=cK0YOGJ58a0

  36. Confidence by The+Living+Fractal · · Score: 1

    What I want to know is this:

    The machine will probably be able to come up with an answer (maybe not the right one) much faster than all of the human opponents. But, what confidence will it have in that answer, and will it realize that a wrong answer will cost it?

    Obviously if the machine just answers immediately (and no 'confidence' factor is involved) then it could provide wrong answers very quickly, and thus just lose money on every question as it "presses the button" to answer the question before the opponents, but answers incorrectly.

    So, IBM, how are you giving Watson a confidence factor? Will Watson's confidence change based on the number of incorrect/correct answers it has given in a row, based on how much time it waited to find the best, most 'confident' answer? In short, will Watson learn?

    --
    I do not respond to cowards. Especially anonymous ones.
  37. Re:I want to see how it responds to this question. by RealGrouchy · · Score: 1

    (for the lazy mods who blindly modded the parent up)

    The real question, is what is the real question for which 42 is the answer?

    And that is why *this* computer's answer would be interesting, because it's designed for Jeopardy, where answers must be in the form of a question.

    I suggest we build a planet, who's sole purpose is to calculate that question...

    The Earth, in the Douglas Adams universe, was NOT a planet; it was an organic supercomputer frequently mistaken for one.

    - RG>

    --
    Hey pal, this isn't a pleasantforest, so don't waste my time with pleasantries!
  38. Re:Why is "Watson" such a popular choice of name? by Bob3141592 · · Score: 1

    I'd presumed that Watson refered to the assistant of Bell who first understood electric speech. Probably a wink to the quirky and confounding associations Jeapordy delights in.

    --
    In theory, there's no difference between theory and practice. In practice, there is.
  39. Alex is a chump. by funwithBSD · · Score: 1

    Let's see if it can Win Ben Steins Money.

    --
    Never answer an anonymous letter. - Yogi Berra
  40. Re:I want to see how it responds to this question. by kodomo · · Score: 1

    better yet. How to reverse the direction of entropy?

  41. Re:Ben Stein is a chump. by wjousts · · Score: 1

    Seriously, he lets himself get taken for a ride by the IDer's. What a moron.

  42. Re:WOW by chuck · · Score: 1

    Um, Trebek doesn't compete. He's the host of the show. The summary is stupid. The computer will be playing against other human contestants.

    --
    My Freakin Blog
  43. Re:I want to see how it responds to this question. by Bigjeff5 · · Score: 1

    Wow. Seriously?

    Ok, first, it was a joke, don't have a cow man.

    Second, I have no clue why it was modded insightful, but whatever.

    Third, no, phrasing answers in the form of a question is easy, have you evern watched Jeopardy before? That's the easiest part of the whole thing, "What is/are X" and "Who is/are X" are all you have to do the answers to phrase them as a question. No, the interesting thing will be if IBM's machine can parse the question in the correct context and come up with the correct answer faster than a human. It's an area that computers thus far have very much sucked at.

    Fourth, wth are you smoking? Do you really take D. Adams that seriously? His whole series was a big joke! Literally!

    And last, yes, the Earth was a planet in Adams' universe. Look up the definition of a planet man. Seriously. Being an organic computer would not negate its planet status. It is a planet because of its size and the fact that it is the largest object in its orbital path. Not being a planet because it was a giant computer doesn't even make sense. One of the questions we ask about whether an object was a planet was NOT "Well, is it a giant computer? Or no?"

    Good night man, are you a professional buzz kill or something? Or is it just something you do as a hobby?

    --
    Security is mostly a superstition... Avoiding danger is no safer in the long run than outright exposure. - Helen Keller
  44. Re:The question is... by silent_artichoke · · Score: 1

    Nasty hobbitses! It cheats, Precious!

  45. Re:Ben Stein is a chump. by The+Archon+V2.0 · · Score: 1

    Seriously, he lets himself get taken for a ride by the IDer's. What a moron.

    WHAT? Richard Nixon's speechwriter isn't a bubbling font of integrity, wisdom, and truth? MY WORLD VIEW HAS BEEN SHATTERED! MY PARADIGM HAS BEEN PERILOUSLY SHIFTED! I CAN'T GO ON!

  46. Re:Game show winnings? by Garrett+Fox · · Score: 1

    A good poker AI might actually be more interesting than a Jeopardy AI, even if the game it played was online so as to eliminate the factor of reading body language. For Jeopardy questions that boil down to "What is the capital of X" or "In what year did X happen?", a winning AI could basically just be Google running on an internal database. In contrast, winning at poker would involve social reasoning about questions like "When this guy suddenly raises his bet, is he often bluffing, and how likely is it that he thinks I think he's bluffing?"

    Build a great Jeopardy AI, and you have a slight upgrade to Google. Build a great poker AI, and you have something that can start to tackle other human social situations.

    Reality check: how good are actual poker AIs? A quick search turns up claims that some are pretty good.

    --
    Revive the Constitution.
  47. won't work by jtgd · · Score: 1

    The 'Watson' computer system uses natural-language processing techniques to break down questions into their structural components and then search its database for relevant answers.

    Well DUH, that won't work! On Jeopardy they give you the answer and you have to respond with the question. Geez, haven't they ever watched the show?

    --
    J