Real-World Outcomes Predicted Using Social Media 93
Hugh Pickens writes "Kevin Kelly writes that researchers at the Social Computing Lab at HP Labs in Palo Alto have found that social media content can predict real world outcomes. In their study, the researchers built a model that used chatter from Twitter to predict accurately the box-office revenues of upcoming movies weeks before the movies were released. When the sentiment of the tweet was factored in (how favorable it was toward the new movie), the prediction was even more exact. To quantify the sentiments in 3 million tweets, the team used anonymous workers from Amazon's Mechanical Turk to rate a sample of tweets, and then trained an algorithmic classifier to derive a rating for the rest. But predicting box office receipts may be only the beginning. 'This method can be extended to a large panoply of topics [PDF], ranging from the future rating of products to agenda setting and election outcomes,' the researchers write. 'At a deeper level, this work shows how social media expresses a collective wisdom which, when properly tapped, can yield an extremely powerful and accurate indicator of future outcomes.'"
This is great for one thing: (Score:3, Funny)
Predicting what a bunch of assholes who use twitter will do.
Re: (Score:2)
Re: (Score:3, Funny)
Enough is enough! I have had it with these motherfucking twitters on this motherfucking web! - Samuel L. Jackson
Re: (Score:2)
And considering that the majority of the movie going public are assholes, it's perfect for predicting box office revenue--or any number of other questions which can be answered by large samples of asshole behavior.
Re: (Score:2)
Psychohistory! (Score:5, Insightful)
These social network predictions were already predicted by the late Mr. Asimov. ;)
Re: (Score:1)
These social network predictions were already predicted by the late Mr. Asimov. ;)
10 points!!! Exactly!
Sounds like they re-discovered the Delphi method (Score:5, Informative)
Wikipedia says:
The Delphi method [wikipedia.org] is a systematic, interactive forecasting method which relies on a panel of experts.
Of course, in this case the "experts" are the movie-going public, who know more about their tastes in movies that anyone is Hollywood. The Delphi method depends on large panels, and n this case th researchers are using large panels indeed. Finally, the iteration is provided by the later tweeters reading earlier tweets before they post.
Re: (Score:1)
Re: (Score:1)
Delphi pools in John Brunners "the shockwave rider" predates asimov.
Only if Brunner also included a time machine.
Predictions (Score:5, Informative)
Re: (Score:2)
Generally 1-5 cents per answer. Most take no more than 2 minutes. So I really quality turker could make $1.50 or more an hour.
Nope, that's not the point (Score:5, Informative)
Putting the fact that increasing sample size does not necessarily increase the power of a predictor, you apparently didn't get the point of their method.
So method A was to simply "grep RamboIX" in these 3 million tweets. That alone already correlated to the box office outcome. However, that also catches messages like "RamboIX suxx, no way I'm going to see or even download this".
So method B was to use machine learning algorithms, combined with some initial work by human drones, to assign a degree of "positiveness" to each message about RamboIX.
While this has nothing to do with increasing sample size, it took the accuracy of the prediction to a whole new level.
I for one think this is a pretty great idea.
Re: (Score:1)
They worked out that comments, essentially, have a sign and counted them appropriately.
I'm not exactly expecting a postcard from Stockholm.
Re: (Score:2)
Actually it's fascinating because the sample size was NOT in itself responsible for the accuracy of the prediction.
Note to self... (Score:2, Interesting)
Note to advertisers (Score:2)
Confirms what I've seen: The Canary Effect (Score:2)
I watch Social Media mentions of things I care about very closely. I've explained to others how I've come to realize there is a definite "canary effect" with the mass sentiment seen via real-time opinion/view venues such as Twitter.
In fact, for items related to "down time" of sites people are routinely faster at registering their dismay at a service being unavailable than expensive site monitoring tools. This isn't exactly predicting future outcomes, but it is an "early warning" indicator that businesses sh
Re:Confirms what I've seen: The Canary Effect (Score:4, Insightful)
Re: (Score:2)
They are already trying their best, competitions to retweet this etc. But I think volume is their enemy here.
Even something as simple as only taking unique messages into consideration would counteract marketing tactics. I am sure there would be an arms race with regard to this.
But a question is why would marketing people bother. It is easy to see why they like their message to be a trending topic, but what is there to gain from gaming some statisticians results.
Re: (Score:2)
Exactly, there is nothing to be gained by trying to game this to make a false prediction, marketer's who know what they are doing will love it's for data mining potential. eg: which new movie or TV show gives the best advertising bang per buck, who should we pay to wear our shoes, etc.
Re: (Score:2)
Uh.. foiling other people's ability to gain an advantage seems like a perfectly good reason to game a system to me.
Any system can be gamed. Any system that can provide or deny an advantage will be gamed.
Re: (Score:2)
Re: (Score:2)
Well.. you're obviously thinking the cost is a lot higher than I am. And it isn't wasting their marketing budget.
Leading up to a product launch, say twitter is used to predict sales. Sales to which a company is trying to match production. Poison twitter data, and you can induce a competitor to overproduce. Which leaves the product looking like it isn't selling. So not only have they spent on marketing, but they are also spending on superfluous production capacity, tying up working capital buying back invent
Re: (Score:2)
Re: (Score:2)
Are people tweeting about this movie because they're excited about it or are they excited about it because people are tweeting about it? The crowds are fickle and easily swayed, especially by themselves.
Re: (Score:2)
I guess it doesn't really matter, because they are excited about it either way.
Re:Confirms what I've seen: The Canary Effect (Score:4, Insightful)
It works until marketing departments at big companies start gaming it in a big way.
No need to game the system when you've already gamed the users! How do you think all these twitterers know to talk about movies before they're released?
Re: (Score:1)
Re: (Score:2)
Re: (Score:2)
Look at Apple, if you want to see the real kings of this viral marketing techniques!
I swear that there was not a single day since months, where there wasn’t some “article” that conveniently contained some Apple product name. Even for simple things like “listening to your mp3 player” which becomes “listening to your iPod”, etc.
I know, because a spam filter in my RSS reader filters them out and shows me statistics about it.
I mean making people think the iPhone would e
Re: (Score:2)
There's only so much you can game. Sure, there'll be a "bump" in gaming at first once the marketeers figure it out, but as soon as the scientists notice this, we'll see bayesian type filtering looking for the drone accounts.
Sure, bots can fool some people, and there will eventually be some good ones. But it's just like spam: there are a lot of approaches to stop them at the gate, and if they get past it, most of them aren't that bright or humanly inconsistent.
I can't imagine that twitter won't have built-in
Re: (Score:2)
Re: (Score:1)
In other words... (Score:3, Insightful)
At a deeper level, this work shows how social media expresses a collective wisdom which, when properly tapped, can yield an extremely powerful and accurate indicator of future outcomes.
In other words - Never underestimate the power of stupid people in large groups.
Re: (Score:1)
Maybe a good prediction for public opinion (Score:3, Interesting)
Re:Maybe a good prediction for public opinion (Score:5, Informative)
However, do we really want to always be driven by public opinion?
I wasn't sure - so I checked twitter and it turns out that we do.
Is it the content or the buzz? (Score:3, Interesting)
Re: (Score:1)
The buzz is important, but it is fickle, a bad movie might launch with high expectations, but people will quickly figure it out.
Re: (Score:2)
yeah... but people went to see Avatar based on buzz, even though the movie thoroughly, totally, absolutely, entirely, wholly, fully, quite, altogether, one hundred percent, downright, outright, in all respects, unconditionally, perfectly, really, to the hilt, to the core, utterly, positively, indisputably, indubitably, beer, unquestionably, beyond any doubt, beyond any question, incontrovertibly, incontestable, irrefutable, unassailably; certainly, surely, definitely, positively, conclusively, plainly, obvi
Psychohistory? (Score:3, Insightful)
Is this an early experiment in the development of psychohistory [wikipedia.org]?
Hari Seldon would be proud.
Panopoly... (Score:1)
Re: (Score:1)
-1 Fail (Score:1)
Here comes the astroturf (Score:2)
The true demise of twitter will be / is when the PR firms that try to take advantage of this flood it with spam, or worse yet, pay people to hype their junk.
Onwards to the next social networking platform!!! I want something with pub/private encryption, non-repudiation, recall, key escrow, supports live pictures, movies, sound, and sound effects, multi-threaded conversations, geolocation, rankings, tagging, filtering, and stuff (yes, I know I contradicted myself a few times, laugh)
Re: (Score:2)
Sounds like Google Wave to me! A couple of plugins and you're sorted.
Re: (Score:3, Informative)
When it happens??? It's been happening from day 1. They're just so good at it that you haven't noticed.
Re: (Score:1)
Re: (Score:1)
Congratulations?
Summary Mistake.. (Score:1, Insightful)
"'At a deeper level, this work shows how social media expresses a collective wisdom which, when properly manipulated, can yield extremely powerful and profitable future income..'"
- Fixed That For You.
So, let me get this straight... (Score:2)
Holy Crap stop the presses! We just invented the Oracle of Delphi! It's all so clear now. The Greeks weren't talking to the gods, but they were talking to a complicated trend analysis computer that tapped into their far reaching social networks!
=P
Nah, in all seriousness though, it's a pretty interesting read.
Re: (Score:2)
Nah, it's more of the Councilor Troi "sensing the obvious and predicting the present" sort of thing. =)
Re: (Score:2)
So let me get this straight. A research institution came to the conclusion that popular things tended to earn a lot of money, while unpopular things tended to tank?
No... a research institution found a way to quickly quantify popularity without expensive market research and focus groups. *That* is the innovation here.
Those terms should not be paired. (Score:1)
Keep Joss Going (Score:1)
Re: (Score:1, Flamebait)
Yes, just what we need, more idiots talking about how 'deep' his shows are. How great he is for 'addressing moral issues'.
I don't think I can take another Buffy. FireFly was fine right up till Serenity when it just turned into to much of 'moral' story than a space western.
Re: (Score:1, Insightful)
Yes, just what we need, more idiots talking about how 'deep' his shows are. How great he is for 'addressing moral issues'.
I don't think I can take another Buffy. FireFly was fine right up till Serenity when it just turned into to much of 'moral' story than a space western.
Heh...you only realized that they were addressing moral concepts in Firefly when the movie came out, and we're the idiots? From the very start that show has been about how Big Government is evil.
effect of viral marketing (Score:1)
Watch out Wall St (Score:2)
I can see it already. Someone in Wall St is look at this thinking they can automate their buying and selling of stocks and beat the market, then they'll say they aren't making enough so they'll create a derivatives market in social media futures, then they'll fiddle things slightly to improve it, then when the bubble bursts (again) we'll look back at this and wonder why it happens again and again.
Re: (Score:2)
Uhm, CNN is about the best indicator of the stock market you can get. Watch CNN long enough and you can predict the markets in general trends, probably rarely will you get news soon enough to make money on any single companies stock.
Broadcast news controls people FAR more than they realize. The difference is, broadcast news sets the tone and people then act on it. Social media just makes those actions apparent in a new place. They were always apparent if you bothered to look.
Had the broadcast news chann
Not if the mpaa has any say (Score:1)
http://www.bloomberg.com/apps/news?pid=20601204&sid=aBtiaTy.Q1vw [bloomberg.com]
"The Motion Picture Association of America asked regulators to reject proposals from two planned exchanges that would allow investors to trade in movie futures.
Approving movie futures contracts would be the "economic equivalent of legalized gambling," MPAA interim Chief Executive Officer Bob Pisano said in a letter March 23 to the Commodity Futures Trading Commission.
The Cantor Futures Exchange, a unit of Cantor Fitzgerald LP, and Trend Exc
Collective wisdom? (Score:4, Funny)
No kidding (Score:4, Insightful)
The quote in the article is such crap:
At a deeper level, this work shows how social media expresses a collective wisdom which, when properly tapped, can yield an extremely powerful and accurate indicator of future outcomes.
No it doesn't show that at all. It shows that what is popular in twitter is popular in the real world. In other words, it shows that twitter is close enough to a representative sample of the general population for many practical purposes. That is all. It doesn't have anything to do with collective wisdom, nor does it help you predict any outcome unless it is primarily dependent on popular opinion.
No shit sherlock (Score:3, Insightful)
Yes, 'the crowd' generally has a pretty good idea of what 'the crowd' is going to do.
If you ask me if I'm going to go see a movie or not than my answer is probably going to pretty accurately reflect what I'm actually going to do.
I'm not exactly sure why this is surprising? Marketers have been doing this for years. They announce products that haven't even hit the drawing board yet. If they get a good response in the form of inquires and other types of interest, they build the device. If they don't then it will just go away.
I think I can simplify this even more (Score:4, Insightful)
Forget the mechanical turk. Just tell me the marketing budget of a movie and how many screens it's going to open on and I'll give you an estimate of the take. It's a pretty strong correlation on large-release movies.
Hollywood has worked so hard to remove the actual quality of the movie from the equation. Get the movie onto a lot of screens early and spend a lot on advertising. Get people in to see it on the first 3 days (Fri, Sat Sun, or Wed-Sun in some cases) before any info about the movie that you don't control (i.e. other than promos) gets out.
Many movies make around half their total theatrical take in the first weekend of release.
Re: (Score:2)
Hollywood doesn't push movies that actually suck. The plots may be generic, and made for mass appeal, but the truly awful releases are not previewed or advertised beyond what it takes to satisfy the egos that made the trash in the first place.
Transformers (I didn't see the second one) was a steaming pile of shit in terms of art, but it was an okay movie. Explosions. Sweaty young women. One dimensional characters. Sounds a lot like the James Bond flicks from the 60s, doesn't it?
Re: (Score:2)
Re: (Score:2)
Has it shown a propensity to predict those? I would suggest their system only measures public awareness which is driven by marketing. It won't find "There's Something About Mary" which took 9 weeks to reach #1 and made 60% as much in its 8th week as in its first.
http://boxofficemojo.com/movies/?page=weekend&id=somethingaboutmary.htm [boxofficemojo.com]
Compare that to "The Hangover", an excellent movie in the same category. A movie anyone in Hollywood would be glad to have made. It made 13% as much in its 8th week as its fi
Fallibility? (Score:1)
Collective wisdom? (Score:1)
These days I would consider that an oxymoron..
This is EASY! (Score:2)
Here's a prediction based upon social networks:
People will become increasingly self-absorbed and focused on triviality and consumption.
During this Holy Week, I think we should all remember that when you post on Facebook or Twitter, it makes Jesus cry.
And it makes Buddha do the "finger down your throat" sign for puking.
Profit and Political Motives Only (Score:2)
But predicting box office receipts may be only the beginning. 'This method can be extended to a large panoply of topics, ranging from the future rating of products to agenda setting and election outcomes,' the researchers write. 'At a deeper level, this work shows how social media expresses a collective wisdom which, when properly tapped, can yield an extremely powerful and accurate indicator of future outcomes.'
Hey, how about using this for something that's actually useful, like predicting and preventing attempts at suicide?
Polling! (Score:3, Insightful)
It beats online polls! (Score:2)
The only difference between this and regular polls is that this is less scientific (since they make no effort to find a random selection of people from the population). It's probably a little better than online polls (probably less manipulation) and a little worse than scientifically-designed polls.
Judge for yourself, here's what they say:
Surprisingly, we discovered that the chatter of a community can indeed be used to make quantitative predictions that outperform those of artificial markets. These information markets generally involve the trading of state-contingent securities, and if large enough and properly designed, they are usually more accurate than other techniques for extracting diffuse information, such as surveys and opinions polls.
So twitter-reading beats markets which beats polls.
Also, what do you mean by "scientifically"? If you mean "like scientists", could you please explain to me what the important properties of what scientists do are?
I have a sad (Score:1)
It doesn't surprise me that a few HP geeks are doing arithmetic analysis of what's supposed to be art. No doubt this model will be used to bang out a bunch of crap optimized for it, which will then disprove the model. It would not surprise me if a niche industry in punitry alts didn't spring up so that people could blog every possible permutation of like/dislike early and so market the winning alts as "market drivers" that could be marketed to studios. Why not? We're already doing that here on slashdot a
Subject vs. Object (Score:1)
The Large Hard-on Collider (Score:1)
Webbots (Score:1)