Could Computers Predict Political Unrest Like They Predict The Weather?

from the doubtful dept

Via betajames, we learn of some research being done on a computer system that's trying to predict political unrest by looking at press coverage.
In a recently published paper, Culturomics 2.0: Forecasting Large-scale Human Behavior Using Global News Media Tone in Time and Space, [Kalev Leetaru] shows that by feeding millions of news articles from around the world into an SGI supercomputer, you can analyze the tone of media coverage and pinpoint moments of unrest: a revolution, riots, or even when a despotic ruler will relinquish power and flee his country.
Of course, so far, the system has only been used on past events, and as it says on the box: past results are no guarantee on future performance. Separately, looking at the examples in the article, it sounds like it notices stuff... well... about the same time a human paying attention would notice stuff. It talks about how right before Pearl Harbor the Japanese press dropped all talk of peace and ramped up criticism of the US. It also discusses Mubarak leaving office and how the press turned really negative right before that. But it wasn't like you needed a computer to tell you that the protests were having an impact and likely leading to the end of Mubarak's reign.

The project sounds interesting, but it seems like Leetaru is overly optimistic about what it'll show. Until there's evidence of it picking out things that you couldn't pick out yourself just by reading the news, I'm not sure there's really any breakthrough.
Hide this

Thank you for reading this Techdirt post. With so many things competing for everyone’s attention these days, we really appreciate you giving us your time. We work hard every day to put quality content out there for our community.

Techdirt is one of the few remaining truly independent media outlets. We do not have a giant corporation behind us, and we rely heavily on our community to support us, in an age when advertisers are increasingly uninterested in sponsoring small, independent sites — especially a site like ours that is unwilling to pull punches in its reporting and analysis.

While other websites have resorted to paywalls, registration requirements, and increasingly annoying/intrusive advertising, we have always kept Techdirt open and available to anyone. But in order to continue doing so, we need your support. We offer a variety of ways for our readers to support us, from direct donations to special subscriptions and cool merchandise — and every little bit helps. Thank you.

–The Techdirt Team

Filed Under: computers, political unrest, predictions


Reader Comments

Subscribe: RSS

View by: Time | Thread


  • icon
    tatheg (profile), 30 Sep 2011 @ 3:41am

    No worries

    If it follows twitter as well as the Main Stream Media it may get lucky from time to time. What was that saying? Even a blind squirrel finds a nut from time to time.

    link to this | view in chronology ]

  • identicon
    Yukitaro, 30 Sep 2011 @ 3:50am

    Psychohistory what?

    I hope Asimov gets proper credit.

    link to this | view in chronology ]

  • identicon
    jimbo, 30 Sep 2011 @ 4:16am

    just tell me when these arse hole entertainment industries will get some brains and actually use them. that will do for a start!

    link to this | view in chronology ]

  • icon
    EveningStarNM (profile), 30 Sep 2011 @ 4:19am

    It's not like it takes any intelligence

    Anyone who can read the unemployment rate or compare average incomes in percentiles can predict political unrest. It might be a little bit harder to pin down when people will rebel against a dictator, but sudden changes in economic conditions that increase poverty ALWAYS precede political unrest. It was true in Tunisia, it was true in Egypt, it was true in Greece, it was true in the rest of Europe, and it's true here in the U.S.

    But perhaps the morons in the U.S. government need a computer to tell them what everybody else already knows: THEY'RE DOING IT WRONG.

    link to this | view in chronology ]

    • identicon
      Anonymous Coward, 30 Sep 2011 @ 6:13am

      Re: It's not like it takes any intelligence

      "Could Computers Predict Political Unrest Like They Predict The Weather?"

      Probably. Computers don't seem to do that good a job at predicting the weather, I don't imagine they can do much worse at predicting political unrest.

      "Anyone who can read the unemployment rate or compare average incomes in percentiles can predict political unrest."

      Sounds like a better political unrest indicator than what we currently have to predict the weather.

      link to this | view in chronology ]

      • icon
        EveningStarNM (profile), 30 Sep 2011 @ 8:23am

        Re: Re: It's not like it takes any intelligence

        It doesn't seem to matter any way. Do you remember this?

        "Red sky in morning, sailor's warning. Red sky at night, sailor's delight."

        Ignoring for the moment that the sky is always red now, the fact that computers can now reliably predict the weather four to five times farther in advance than that old adage might indicate that computer predictions of civil unrest, if similar improvements were made, could be useful.

        Unfortunately, we already know that it won't. For instance, we've known for 100 years that current conservative economic policies are bad for the country, and even though we threw Republicans out of government in 2008, we're still following conservative policies today. We knew 80 years ago that prohibition would lead to a huge and expensive increase in violent crime. We knew 60 years ago that increases in atmospheric pollution would result in higher and expensive increases in temperature. We knew 40 years ago that plastics and many unnatural chemicals would build up the bodies of every human being on the planet.

        The most accurate prediction is of no use if we choose to ignore it.

        link to this | view in chronology ]

        • icon
          Butcherer79 (profile), 30 Sep 2011 @ 12:03pm

          Re: Re: Re: It's not like it takes any intelligence

          Shepherds warning/Delight where I come from

          Or:
          Red sky at night, shepherds house on fire...

          link to this | view in chronology ]

    • icon
      The eejit (profile), 30 Sep 2011 @ 7:28am

      Re: It's not like it takes any intelligence

      To be fair, most of the current corps of politicians could probably be safely replaced with AI equivalents.

      link to this | view in chronology ]

  • icon
    Tim K (profile), 30 Sep 2011 @ 5:02am

    Perhaps the advantage of the computer is that it can 'read' thousands of articles whereas a person can read significantly less. Obviously, the more articles read the more valid the conclusions of the reader are. If I read 5 articles about the goings on in Saudi Arabia I may draw different conclusions than a computer that read 65,000 articles about Saudi Arabia.

    I'm not saying that the computer is going to be a huge asset, but I think its wrong to discount it simply because a person can read articles and draw conclusions too. I can do math too, but a computer is faster and better at it.

    link to this | view in chronology ]

  • icon
    Richard (profile), 30 Sep 2011 @ 5:03am

    The very existence

    The very existence of a computer program that could predict political unrest would itself have an impact on the unrest.

    This could work two ways - it could become a self fulfilling prophecy - once people hear that the program is predicting revolt in their country - they have the confidence to revolt.

    Alternatively the authorities make use of the predict to somehow attempt to avert it ... not holding my breathe for that one though.

    link to this | view in chronology ]

  • icon
    NegruVoda (profile), 30 Sep 2011 @ 5:09am

    Re:Could Computers Predict Political Unrest Like They Predict The Weather?

    You mean badly? They probably could.

    link to this | view in chronology ]

  • identicon
    Anonymous Coward, 30 Sep 2011 @ 5:37am

    One person can reach wrong conclusions for looking at just part of the whole, the thing about computers is that they don't have bias, they don't exclude themselves from data, that may be more reliable that trusting a person unless one have proved himself to be good at controlling his emotions while doing analysis.

    link to this | view in chronology ]

    • icon
      Josh in CharlotteNC (profile), 30 Sep 2011 @ 7:19am

      Re:

      Computers have the bias that the programmers built into them. If it's some sort of pseudo-neural network learning system (ie Watson), then it's biased by the previous data it incorporated, plus any bias that was built in.

      Second rule of tech support: computers always do exactly what you tell them to do, not what you mean for them to do.

      link to this | view in chronology ]

      • identicon
        Anonymous Coward, 30 Sep 2011 @ 2:09pm

        Re: Re:

        Very true.
        Still, they do tend to be less biased despite all those problems.

        And I believe in happy accidents where people biased towards one thing are trying to prove/find another and discover something about something unrelated to their original quest.

        More or like Teflon was discovered, when Roy Plunkett was trying to make a better refrigerant and discovered Teflon.

        http://en.wikipedia.org/wiki/Polytetrafluoroethylene

        And I know you can approximate moods analysing the tone of the words used just by counting the words used in posts everywhere.

        http://en.wikipedia.org/wiki/Frequency_analysis

        You also can infer level of education from a native speaker of a language, age, sex, some psychological traits and some other parameters that although never exactly right always puts you on the ballpark of where that individual is.

        Those things are possible or mothers everywhere wouldn't be able to predict how their sons and daughters would react.

        You can count other things too, like body expressions, patterns emerge in social structures, but humans are very bad in counting and interpreting those, maybe because we are not trained or learn to do that, but it doesn't mean those patterns are not there.

        link to this | view in chronology ]

  • identicon
    Anonymous Coward, 30 Sep 2011 @ 7:34am

    I have have no confidence that this effort will ever reliably forecast anything. An obvious problem being how to effectively accounting for media bias.
    For example, the mainstream media here in the US have largely ignored the Occupy Wall Street demonstrations in which thousands of people have been massed in New York's financial district in opposition to Wall Street's corruption for the last week and a half. Thousands of people a week and a half, and still the US corporate owned media has very little – if any coverage. (None – print or video – for the first week.) Now consider the coverage given to Tea Party protesters who's messages are in alignment with corporatist goals (if not, in fact, authored by them).
    So, if a significant social trend/narrative is not reported (or is dramatically under reported) and its converse enjoys heavy coverage, how do you program to correct for the biases in the data sample without introducing yet other biases?
    The cardinal rule of computing: GIGO (garbage in – garbage out).

    link to this | view in chronology ]

    • identicon
      Anonymous Coward, 30 Sep 2011 @ 8:10am

      We need to co-mingle theories... Since the 'infinite monkey' theory was just proved (somewhat), I propose the INFINITE GARBAGE theorum

      I state that if we dump enough garbage into a computer system, eventually an open, collaborative, communicative, people centered, responsible government model will emerge....

      Or perhaps the Infinite Monkeys will just fly out of my butt (86.7% more likely than the above, of course 68% of statistics are made up on the spot, like these)....

      You never know what might happen with these 'infinite' theories.....

      link to this | view in chronology ]

      • icon
        EveningStarNM (profile), 30 Sep 2011 @ 8:31am

        Re: Infinite Monkey Theory

        I trust you more than I trust the software engineer's model because you admitted that your statistics were lies.

        link to this | view in chronology ]

    • identicon
      Anonymous Dissenter, 30 Sep 2011 @ 8:12am

      Re:

      Thank you, you saved me the time of typing that same sentiment.

      link to this | view in chronology ]

  • icon
    Butcherer79 (profile), 30 Sep 2011 @ 8:06am

    If it predicts political unrest as well as my local weather, prepare for some er, errors, mass panic.
    I have a question for debate:
    Would the prediction of political unrest go some way to causing it?

    link to this | view in chronology ]

    • icon
      EveningStarNM (profile), 30 Sep 2011 @ 8:47am

      Re: The Thomas theorem

      In a word, "yes". The theory of self-fulfilling prophecies actually is a corollary to the observer effect. These predictions could even lead to that unrest becoming worse than it otherwise would.

      For instance, let's say you're a dictator, and your computer tells you that the people will rebel next year. So you beef up your army and clamp down on civilians. As a result, the people will get angrier faster.

      link to this | view in chronology ]

      • icon
        Butcherer79 (profile), 30 Sep 2011 @ 12:09pm

        Re: Re: The Thomas theorem

        interesting though, don't you think? I don't think you could have such a system without it at least contributing to the unrest it predicts

        link to this | view in chronology ]

        • icon
          EveningStarNM (profile), 30 Sep 2011 @ 1:19pm

          Re: Re: Re: The Thomas theorem

          Yes, and you just KNOW that we'll use such a system knowing that it will make the problem worse. I also sometimes wonder what our species would be like if puberty came a little later, when we were more mentally mature.

          It's a pity that wishful thinking is my strongest suit.

          link to this | view in chronology ]

  • identicon
    Muad'Dib, 30 Sep 2011 @ 8:34am

    Who needs a computer to predict future unrest when you have a mentat?

    It needs more than reason, it needs imagination. Societal undertones are not enough to extrapolate unrest, one also needs to analyze the motives of those who lead. It must operate under the assumption that nothing is done if not for political gain, whether for the overly oppressed or the tactful aristocrats.

    You can expect better results from Paul, the Octopus.

    link to this | view in chronology ]

    • identicon
      Anonymous Coward, 30 Sep 2011 @ 2:21pm

      Re: Who needs a computer to predict future unrest when you have a mentat?

      I don't believe that is true, I think society is made up of individual that have a limited range of reactions to stimuli and finding out what stimuli triggers certain reactions is not that hard, one just need to pay attention even if not a 100% it can still be useful.

      Just like the weather if you see a storm in the ocean heading your way even if for some reason it dissipates you still take precautions because you don't know if it will reach you or not, but there is a chance that it will and the more close in time the storm gets the more difficult is for things to change suddenly, anger don't just disappear, love just don't disappear it takes some time.

      By analysing what how people signal things you can have a somewhat accurate picture of how they will react based on their on external cues for which language is just one.

      link to this | view in chronology ]

      • identicon
        Muad'Dib, 1 Oct 2011 @ 6:26pm

        Re: Re: Who needs a computer to predict future unrest when you have a mentat?

        Ah, because humanity is in a finite system, human reactions are finite and consistent, thus predictable. You say that a certain stimuli triggers a certain reaction in every person, leaving a trail that one can seek. How right you are! But when that is compounded with people among people, the number of possibilities is nears infinity, and the prime computation will require more than reason. Governments may rise and fall for reasons which appear insignificant. What small events! An argument between two women... which way the wind blows on a certain day... a sneeze, a cough, the length of a garment or the chance collision of a fleck of sand and a courtier's eye.

        You must reduce your strategy to its point of application. Where does one apply strategy? At a particular place and with a particular people in mind. But even with the greatest concern for minutiae, some small detail with no significance attached to it will escape you. Can your strategy be reduced to the ambitions of a regional governor's wife?

        link to this | view in chronology ]

  • identicon
    Gill Bates, 30 Sep 2011 @ 8:39am

    We do nat need that

    Weather prediction is wrong about 60% of the time.

    DO we REALLY want to "predict" political unrest that badly?

    link to this | view in chronology ]

  • identicon
    Anonymous Coward, 30 Sep 2011 @ 9:48am

    Wait...wait. So a computer can now read the news and interpret it as well as a human and we're not excited? Not even a little?

    link to this | view in chronology ]

  • icon
    hmm (profile), 30 Sep 2011 @ 11:36am

    no need

    Weather prediction is slightly MORE accurate than the MPAA/RIAA predictions of a successful business model.....

    If the MAFIAA were in charge of TV weather reports, we'd all be long dead from floods before they thought to issue a severe weather warning.

    In other news - man arrested for standing outside in the rain....RIAA claims getting rained on outside is a public performance by 'The Clouds' and demands a fee.

    link to this | view in chronology ]

  • icon
    Casey-Mahoney-Bradp (profile), 30 Sep 2011 @ 12:18pm

    social Media

    Social Media is very strong in todays world.You can follow the Tweets or the trends on Facebook to see what is going on all over the world.People can group together and form a rally within an hour from using Twitter.

    Thanks Casey Mahoney Bradp

    link to this | view in chronology ]


Follow Techdirt
Essential Reading
Techdirt Deals
Report this ad  |  Hide Techdirt ads
Techdirt Insider Discord

The latest chatter on the Techdirt Insider Discord channel...

Loading...
Recent Stories

This site, like most other sites on the web, uses cookies. For more information, see our privacy policy. Got it
Close

Email This

This feature is only available to registered users. Register or sign in to use it.