Twitter Cannot Predict Elections Either

Claims that Twitter can predict the outcome of elections are riddled with flaws, according to a new analysis of research in this area

It wasn't so long ago that researchers were queuing up to explain Twitter's extraordinary ability to predict the future.  

Tweets, we were told, reflect the sentiments of the people who send them. So it stands to reason that they should hold important clues about the things people intend to do, like buying or selling shares, voting in elections and even about paying to see a movie. 

Indeed various researchers reported that social media can reliably predict the stock market, the results of elections and even box office revenues

But in recent months the mood has begun to change. Just a few weeks ago, we discussed new evidence indicating that this kind of social media is not so good at predicting box office revenues after all. Twitter's predictive crown is clearly slipping. 

Today, Daniel Gayo-Avello, at the University of Oviedo in Spain, knocks the crown off altogether, at least as far as elections are concerned. His unequivocal conclusion: “No, you cannot predict elections with Twitter.”

Gayo-Avello backs up this statement by reviewing the work of researchers who claim to have seen Twitter's predictive power. These claims are riddled with flaws, he says.

For example, the work in this area assumes that all tweets are trustworthy and yet political statements are littered with rumours, propaganda and humour. 

Neither does the research take demographics into account. Tweeters are overwhelmingly likely to be younger and this, of course, will bias any results.   "Social media is not a representative and unbiased sample of the voting population," he says.

Then there is the problem of self selection. The people who make political remarks are those most interested in politics. The silent majority is a huge problem, says Gayo-Avello and more work needs to be done to understand this important group.

Most damning is the lack of a single actual prediction. Every analysis on elections so far has been done after the fact. "I have not found a single paper predicting a future result," says Gayo-Avello.

Clearly, Twitter is not all it has been cracked up to be when it comes to the art of prediction. Given the level of hype surrounding social media, it's not really surprising that the more sensational claims do not stand up to closer scrutiny. Perhaps we should have seen this coming (cough).

Gayo-Avello has a solution. He issues the following challenge to anybody working in this area: "There are elections virtually all the time, thus, if you are claiming you have a prediction method you should predict an election in the future!" 

Ref: arxiv.org/abs/1204.6441: “I Wanted to Predict Elections with Twitter and all I got was this Lousy Paper”: A Balanced Survey on Election Prediction using Twitter Data

View the original article here

Psychologists Use Social Networking Behavior to Predict Personality Type

The ability to automatically determine personality type could change the way social networks target services to users

One of the foundations of modern psychology is that human personality can be described in terms of five different forms of behavior. These are:

1. Agreeableness--being helpful, cooperative and sympathetic towards others
2. Conscientiousness--being disciplined, organized and achievement-oriented 
3. Extraversion--having a higher degree of sociability, assertiveness and talkativeness 
4. Neuroticism--the degree of emotional stability, impulse control and anxiety 
5. Openness--having a strong intellectual curiosity and a preference for novelty and variety

Psychologists have spent much time and many years developing tests that can classify people according to these criteria. 

Today, Shuotian Bai at the Graduate University of Chinese Academy of Sciences in Beijing and a couple of buddies say they have developed an online version of the test that can determine an individual's personality traits from their behavior on a social network such as Facebook or Renren, an increasingly popular Chinese competitor.

Their method is relatively simple. These guys asked just over 200 Chinese students with Renren accounts to complete online, a standard personality test called the Big Five Inventory, which was developed at the University of California, Berkeley during the 1990s.

At the same time, these guys analyzed the Renren pages of each student, recording their age and sex and various aspects of their online behavior such as the frequency of their blog posts as well as the emotional content of the posts such as whether angry, funny or surprised  and so on. 

Finally, they used various number crunching techniques to reveal correlations between the results of the personality tests and the online behavior. 

It turns out, they say, that various online behaviors are a good indicator of personality type. For example, conscientious people are more likely to post asking for help such as a location or e-mail address; a sign of extroversion is an increased use of emoticons; the frequency of status updates correlates with openness; and a measure of neuroticism is the rate at which blog posts attract angry comments.

Based on these correlations, these guys say they can automatically predict personality type simply by looking at an individual's social network statistics. 

That could be extremely useful for social networks. Shuotian and comapny point out that a network might use this to recommend specific services. They give the rather naive example of an outgoing user who may prefer international news and like to make friends with others. 

Other scenarios are at least as likely. For example, such an approach might help to improve recommender systems in general. Perhaps people who share similar personality characteristics are more likely to share similar tastes in books, films or each other. 

There is also the obvious prospect that social networks would use this data for commercial gain; to target specific adverts to users for example. And finally there is the worry that such a technique could be used to identify vulnerable individuals who might be most susceptible to nefarious persuasion.

Ethics aside, there are also certain questions marks over the result. One important caveat is how people's response to psychology studies online differs from those done at other times. That could clearly introduce some bias. Then there are the more general questions of how online and offline behaviours differs and how these tests vary across cultures. These are things that Shuotian and Co. want to study in the future.

In the meantime, it is becoming increasingly clear that the data associated with our online behavior is a rich and valuable source of information about our innermost natures. 

Ref: arxiv.org/abs/1204.4809: Big-Five Personality Prediction Based on User Behaviors at Social Network Sites

View the original article here