The effect of social contexts on network response to emergencies

Abstract

We study how social communication behaviors change when the population encounters unfamiliar conditions, which has potential to provide insights into improving resource management in response to emergencies and crises, as well as to offer new perspective on information propagation. Using anonymous billing records of 10 million mobile phone subscribers in a western European country from 2007 to 2009, we compare call activity in the immediate aftermath of a set of emergencies with scheduled events (such as rock concerts and sporting events). We find that, despite the communication avalanches accompanying large-scale emergencies due to the significant population of eyewitnesses, the calling targets of these calls are more predictable than those in normal situations. We use predictive models to discover key factors related to people's calling decisions and show that people's calling behaviors in emergencies differ from those in normal situations (scheduled events or non-event periods). The finding suggests people's effective social networks change with different situations. Comparing with their social networks in normal period, people's strong ties are reasserted during emergencies, but may adjusted by other context-dependent factors such as geographical distance. In contrast, the networks are more likely to be reshuffled by scheduled activities.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jan 01, 2011
Accession Number
ADA557828

Entities

People

  • David Lazer
  • Yu-Ru Lin

Organizations

  • Harvard University

Tags

Communities of Interest

  • Air Platforms
  • Human Systems

DTIC Thesaurus Topics

  • Abstracts
  • Base Lines
  • Boundaries
  • Communication Networks
  • Emergencies
  • Human Population
  • Information Operations
  • Military Research
  • Mobile Communications
  • Mobile Phones
  • Mobility
  • Networks
  • Predictive Modeling
  • Resource Management
  • Simulations
  • Social Networks

Readers

  • Computer Networking
  • Political Violence and Terrorism Studies.
  • Theoretical Analysis.