A comparative analysis for spatio-temporal spreading patterns of emergency news

Mingjiao Si, Lizhen Cui, Wei Guo, Qingzhong Li, Lei Liu, Xudong Lü, Xin LüScientific 2020 · DOI

Abstract: Understanding how online emergency news propagates is vital for effective disaster management and information dissemination. While previous research often focuses on single events, this study introduces a comprehensive evaluation framework to quantify general spreading patterns across 81 million reposts on Sina Weibo. By comparing emergency news with political, social, tech, and entertainment news, the authors find that emergency information typically follows a "fast and shallow" trajectory. It exhibits shorter life cycles, fewer post-peak fluctuations, and limited network depth. Notably, emergency news is less dependent on high-influence users, suggesting an egalitarian spread driven by information urgency rather than social status.

Introduction

Social media has transformed into a primary conduit for information during crises. However, emergency managers often lack generalizable models for how this information moves through networks, as most academic focus remains on isolated disasters. This research addresses the gap by quantifying the "spatio-temporal" (space and time) spreading patterns of emergency news compared to other topics, providing a blueprint for predicting information reach and duration.

Methods and Data

The study analyzed a massive dataset from Sina Weibo, comprising 81 million repost records from over 3.5 million users across 193 major events in 2018. These events were categorized into five types: Emergency, Political, Social, Tech, and Entertainment.

The researchers evaluated information spread across five dimensions: temporal dynamics, topological structure, propagation modes (visual morphologies), user engagement (efficiency), and information coverage.

Temporal Characteristics: The "One-and-Done" Peak

Emergency news is characterized by its urgency and rapid decay. Unlike entertainment news, which may experience multiple waves of interest as different "influencers" pick up the story, emergency news typically has a single, sharp peak followed by a swift decline.

Temporal spreading patterns for different news types, showing the rapid decay of emergency news (Earthquake) compared to entertainment and social topics.

The data shows that emergency news has the shortest median lifecycle (15 days) and the shortest active period (1–3 days). The mean active period for emergency news is just 1.78 days, significantly lower than the 6.75 days observed for entertainment news.

Topological Patterns and Propagation Modes

The structure of how news travels—its "propagation tree"—reveals the depth of penetration into social networks. Emergency news is remarkably "shallow," rarely exceeding five levels of depth from the original source. In contrast, entertainment news can reach up to nine levels.

The study identified four primary propagation modes based on network morphology: Ripple (direct fans only), Dandelion (weak secondary spread), Colony (multiple small clusters), and Firework (strong, multi-layered dissemination).

Visual morphologies of information propagation: (b) Dandelion, (c) Colony, and (d) Firework modes.

While complex modes like "Firework" are common in entertainment, the vast majority (79%) of emergency news remains in the "Ripple" mode, spreading quickly but not deeply.

The Egalitarian Nature of Emergency Spread

A key finding is that user influence (follower count) matters least during emergencies. While "power users" drive entertainment trends, emergency information is often propelled by the intrinsic value of the news itself. Even users with small followings can trigger massive reposting chains if the information is critical. Furthermore, emergency news boasts high "repost efficiency" (70–80%), indicating that high engagement numbers represent unique individuals rather than a small group of users reposting the same content repeatedly.

Conclusion

By mapping these patterns, the authors developed an ensemble learning model capable of predicting information coverage with high accuracy (R-squared of 0.768). These insights allow emergency managers to better understand the window of opportunity for disseminating vital safety information and highlight the need for strategies to "re-popularize" essential news before it decays. Future work will focus on integrating these patterns with real-time rumor detection to ensure the veracity of rapidly spreading emergency information.

Summary updated

Abstract

Understanding the propagation characteristics of online emergency news communication is of great importance to guiding emergency management and supporting the dissemination of vital information. However, existing methods are limited to the analysis of the dissemination of online information pertaining to a specific disaster event. To study the quantification of the general spreading patterns and unique dynamic evolution of emergency-related information, we build a systematic, comprehensive evaluation framework and apply it to 81 million reposts from Sina Weibo, Chinese largest online microblogging platform, and perform a comparative analysis with four other types of online information (political, social, techs, and entertainment news). We find that the spreading of emergency news generally exhibits a shorter life cycle, a shorter active period, and fewer fluctuations in the aftermath of the peak than other types of news, while propagation is limited to a few steps from the source. Furthermore, compared with other types of news, fewer users tend to repost the same piece of news multiple times, while user influence (which depends on the number of fans) has the least impact on the number of reposts for news of emergencies. These comparative results provide insights that will be useful in the context of disaster relief, emergency management, and other communication path prediction applications.
BibTeX
@inproceedings{si2020comparative,
  author = {Mingjiao Si and Lizhen Cui and Wei Guo and Qingzhong Li and Lei Liu and Xudong Lü and Xin Lü},
  title = {{A comparative analysis for spatio-temporal spreading patterns of emergency news}},
  booktitle = {Scientific},
  year = {2020},
  doi = {10.1038/s41598-020-76162-7},
}
Figure