City-level Geolocation of Tweets for Real-time Visual Analytics

TL;DR

This paper enhances a deep learning model for city-level geolocation of tweets and integrates it into a visual analytics system to improve real-time situational awareness using non-geotagged tweets.

Contribution

It adapts and evaluates a deep learning model for city-level geolocation prediction and integrates it into a real-time visual analytics system.

Findings

Deep learning model outperforms previous methods.

System improves real-time situational awareness.

Model effectively predicts locations of non-geotagged tweets.

Abstract

Real-time tweets can provide useful information on evolving events and situations. Geotagged tweets are especially useful, as they indicate the location of origin and provide geographic context. However, only a small portion of tweets are geotagged, limiting their use for situational awareness. In this paper, we adapt, improve, and evaluate a state-of-the-art deep learning model for city-level geolocation prediction, and integrate it with a visual analytics system tailored for real-time situational awareness. We provide computational evaluations to demonstrate the superiority and utility of our geolocation prediction model within an interactive system.