Geotagging One Hundred Million Twitter Accounts with Total Variation Minimization

Abstract

Geographically annotated social media is extremely valuable for modern information retrieval. However, when researchers can only access publicly-visible data, one quickly finds that social media users rarely publish location information. In this work, we provide a method which can geolocate the overwhelming majority of active Twitter users, independent of their location sharing preferences, using only publicly-visible Twitter data. Our method infers an unknown user's location by examining their friend's locations. We frame the geotagging problem as an optimization over a social network with a total variation-based objective and provide a scalable and distributed algorithm for its solution. Furthermore, we show how a robust estimate of the geographic dispersion of each user's ego network can be used as a per-user accuracy measure which is effective at removing outlying errors. Leave-many-out evaluation shows that our method is able to infer location for 101,846,236 Twitter users at a median error of 6.38 km, allowing us to geotag over 80\% of public tweets.

arXiv comment9 pages, 8 figures, accepted to IEEE BigData 2014, Compton, Ryan, David Jurgens, and David Allen. "Geotagging one hundred million twitter accounts with total variation minimization." Big Data (Big Data), 2014 IEEE International Conference on. IEEE, 2014

References in corpus (6)

• Geo-located Twitter as the proxy for global mobility patterns
• A tale of many cities: universal patterns in human urban mobility
• The Twitter of Babel: Mapping World Languages through Microblogging Platforms
• Parallel Coordinate Descent for L1-Regularized Loss Minimization
• Distance Is Not Dead: Social Interaction and Geographical Distance in the Internet Era
• Parallel Coordinate Descent Methods for Big Data Optimization

Cited by in corpus (17)

• A Survey of Location Prediction on Twitter
• Social Sensing of Floods in the UK
• Using Social Media to Predict the Future: A Systematic Literature Review
• Multiview Deep Learning for Predicting Twitter Users' Location
• Still out there: Modeling and Identifying Russian Troll Accounts on Twitter
• Privacy in Social Media: Identification, Mitigation and Applications
• Modeling Spatiotemporal Factors Associated With Sentiment on Twitter: Synthesis and Suggestions for Improving the Identification of Localized Deviations
• Twitter as a Source of Global Mobility Patterns for Social Good
• Your Actions Tell Where You Are: Uncovering Twitter Users in a Metropolitan Area
• Analyzing the Engagement of Social Relationships During Life Event Shocks in Social Media
• Twitter User Geolocation using Deep Multiview Learning
• Content-Aware Tweet Location Inference using Quadtree Spatial Partitioning and Jaccard-Cosine Word Embedding
• Inferring the geographic focus of online documents from social media sharing patterns
• On the Accuracy of Hyper-local Geotagging of Social Media Content
• Deriving Disinformation Insights from Geolocalized Twitter Callouts
• Using Sociolinguistic Variables to Reveal Changing Attitudes Towards Sexuality and Gender
• Location Inference from Tweets using Grid-based Classification