Skip to Content
Uncategorized

Twitter "Exhaust" Reveals Patterns of Unemployment

Twitter data mining reveals surprising detail about socioeconomic indicators but at a fraction of the cost of traditional data-gathering methods, say computational sociologists.

Human behaviour is closely linked to social and economic status. For example, the way an individual travels round a city is influenced by their job, their income and their lifestyle.

So it shouldn’t come as a surprise that economic status might also be reflected in patterns of social media behaviour. Indeed, that’s exactly what, say Alejandro Llorente at the Autonomous University of Madrid in Spain and a few pals. Today, these guys show that the broad pattern of tweets across cities and counties in Spain reveals fascinating detail about unemployment rates in these areas.

These guys began with a database of 19.6 million geolocated tweets in Spain published between November 2012 and June 2013. Llorente and co wanted to correlate these tweets with regions of economic activity but these are not easy to determine. That’s because they do not correspond well to the administrative boundaries in Spain, which reflect historical and political boundaries rather than economic ones.

So the team analysed the rate at which messages were exchanged between regions using a standard community detection algorithm. This revealed 340 independent areas of economic activity, which largely coincide with other measures of geographic and economic distribution. “This result shows that the mobility detected from geolocated tweets and the communities obtained are a good description of economical areas,” they say.

Finally, they looked at the unemployment figures in each of these regions and then mined their database for correlations with twitter activity.

The results show clear differences between regions with high and low unemployment. For example, the rate of tweeting between 9am and midday on weekdays is significantly higher in areas of high unemployment. These tweets are more likely to contain words such as job or unemployment. And the messages are also more likely to contain spelling mistakes, perhaps reflecting a lower level of education among the unemployed.

“We demonstrate that behavioural features related to unemployment can be recovered from the digital exhaust left by the microblogging network Twitter,” say Llorente and co.

That’s important because this kind of analysis is quick and easy compared to traditional methods of data collection, such as surveys. These are expensive, so much so that some countries have considered abandoning them in times of economic hardship to save money.

The possibility that Twitter data can provide a quick and cheap overview of unemployment is therefore an interesting alternative. What’s more, it allows governments and policy makers to monitor changes in the population, more or less in real time.

“The immediacy of social media may also allow governments to better measure and understand the effect of policies, social changes, natural or man-made disasters in the economical status of cities in almost real-time,” say Llorente and co, adding that their techniques should be applicable anywhere in the world.

Work like this shows how the nature of economic data gathering is changing. It’ll be interesting to see how quickly governments and other organisations adapt.

Ref: arxiv.org/abs/1411.3140 : Social Media Fingerprints Of Unemployment

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

OpenAI teases an amazing new generative video model called Sora

The firm is sharing Sora with a small group of safety testers but the rest of us will have to wait to learn more.

Google’s Gemini is now in everything. Here’s how you can try it out.

Gmail, Docs, and more will now come with Gemini baked in. But Europeans will have to wait before they can download the app.

This baby with a head camera helped teach an AI how kids learn language

A neural network trained on the experiences of a single young child managed to learn one of the core components of language: how to match words to the objects they represent.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.