The unprecedented role of SMS in disaster response

Published 25 February 2010

In Haiti, volunteers set up an SMS messaging system which allow individuals in earthquake-affected areas to text their location and urgent needs in real time for free; since the majority of incoming text messages were in Creole, thousands of volunteers agreed to serve as instant translators

What if, then, we could communicate with disaster affected communities in real-time just days after a major disaster like the quake in Haiti?

This question is posed by Patrick Meier in Haiti Rewired. Meier is the director of Crisis Mapping and Strategic Partnerships at Ushahidi and the co-director of Crisis Mapping and Early Warning at the Harvard Humanitarian Initiative (HHI). He is the co-founder of the International Network of Crisis Mappers (CM*Net) and the International Conference on Crisis Mapping (ICCM) series.

Meier writes that thanks to a partnership between the Emergency Information Service (EIS), InSTEDD, Ushahidi, Haitian Telcos and the U.S. State Department, such communication in real time with disaster affected communities materialized. Just four days after the earthquake, Haitians could text their location and urgent needs to “4636” for free.

I will focus primarily on the way that Ushahidi used 4636. Since the majority of incoming text messages were in Creole, we needed a translation service.

Meier’s colleague Brian Herbert from Ushahidi built a dedicated interface for volunteer translators. He reached out to the Services Employee International Union (SEIU) which mobilized several thousand Haitian volunteers to the cause. Thus, not only was Ushahidi crowd-sourcing crisis information in near real-time, but also crowd-sourcing translation in near real-time.

Meier explains that text messages are translated into English just minutes after they leave a mobile phone in Haiti. The translated messages then appear directly on the Ushahidi platform.

His post contains many good screenshots of SMS messages in the original and in translation.

Meier also notes that on a more “macro” level, he recently reached out to colleagues at the EC’s Joint Research Center (JRC) to leverage their automated sentiment (“mood”) analysis platform. Sentiment Analysis is a branch of natural language processing (NLP) that seeks to quantify positive vs. negative perceptions; similar to “tone” analysis. Meier suggested that they use their platform on the incoming text messages from Haiti to get a general sense of changing mood on an hourly basis. You may be interested in a previous blog post by Meier on the use of Sentiment Analysis for early warning.