Expanding awareness: Comparing location, Keyword, and network filtering methods to collect hyperlocal social media data

Abstract

Opportunities to collect real-time social media data during a crisis remain limited to location and keyword filtering despite the sparsity of geographic metadata and the tendency of keyword-based methods to capture information posted by remote rather than local users. Here we introduce a third, network filtering method that uses social network ties to infer the location of social media users in a geographic community and collect data from networks of these users during a crisis. In this paper we compare all three methods by analysing the distribution of situational reports of infrastructure damage and service disruption across location, keyword, and network-filtered social media data during a weather emergency. We find that network filtering doubles the number of situational reports collected in real-time compared to location and keyword filtering alone, but that all three methods collect unique reports that can support situational awareness of incidents occurring across a community.