Never ending conflicts in the East: Data technologies to keep communities safe

The story of Elva very much parallels the story of a 1960’s garage rock band that exploded to change the world. It all started with a small group of world-traveling, peace-loving, humanitarian conflict prevention professionals in Georgia. The year was 2010; the crisis was the violence and lack of human security that was persisting in the Georgian region of Shida Kartli since the Russian invasion of 2008. There were consistent shootings and other incidents that threatened the security and livelihoods of local communities, yet due to the remoteness of the afflicted region, policy-makers lacked structured, community-driven information about local security needs and perceptions.

This region is the area along the conflict border that divides the disputed territory of South Ossetia from the Georgian region of Shida Kartli. The area is so remote that, even today, there is little to no internet connection. Before Elva, a common way to communicate security incidents among community members was by word-of-mouth. This means of communication risked the spread of rumors and misinformation in the area, thereby contributing to communities’ sense of insecurity. After witnessing these problems, a small group of local NGO workers and conflict transformation experts quickly came to the conclusion that community-wide data-driven incident reporting would beget a powerful solution to the region’s problems. Elva Director Jonne Catshoek explains that they "needed to find an on-line solution to this off-line problem and if they could get the communities to send them reports through SMS” then they could begin addressing the problems at their source.

And thus, Elva was born. It started as a side project between Catshoek, a few of his colleagues, and the conflict transformation organization, Saferworld. They worked during the day to earn money, and worked during the night to create a technology platform that could potentially save many lives. It went on like that until just recently, when Catshoek and his colleagues received support from the United Nations Development Programme (UNDP) to sustain their work on a full-time basis, in cooperation with Saferworld.

The platform allows citizen reporters to submit reports via text message, Twitter, Facebook or email. It houses data analytics and mapping capabilities, similar to those of Ushahidi; however there are three major differences between the platforms:

1. Elva makes use of SMS questionnaires that allow for the collection of answers to up to 40 different questions per SMS message - this provides in-depth and structured baseline data about incidents, implemented policies, and programs in the area.
2. Elva is aimed at facilitating two-way communication between communities and decision-makers.
3. Elva is interested in monitoring long-term developments and needs in target communities, and seeing how these are affected by the implementation of various policies.

Statistical Analyses for Early Warnings on Civil Unrest

With respect to the safety incidents that have been occurring along Georgia’s conflict border with South Ossetia, Elva has been recording and tracking reports about these incidents since 2010. Elva is in the process of developing an early warning system that is based on regressional analysis modeling of this time series data.

The team is fine tuning this model by using the underlying dataset as a baseline from which to identify real-time variables as they occur, and incorporate these instances into the forecasting model. This early warning forecasting model will be used in an attempt to prevent potential instances of violence by alerting authorities so that they can address the “trigger” incidents directly, thereby quashing subsequent violence that is likely to happen further down the timeline.

Engineering a Mass SMS Community Alert System

Elva’s software engineers designed the platform using mostly PHP, Java Script, HTML5, and CSS3. As previously stated, the platform accepts reports via SMS, Twitter, Facebook, and email. It uses MySQL Community Edition as its backend database and has received up to 1,000 text messages per minute during peak reporting intervals. For its mapping application, Elva is built on data from OpenStreetMap and Google Maps. Elva developers have written custom code for visualizing the data on the maps. For its graphs, tables, and other statistical overviews, Elva is built on the Highcharts API.

*Note: Data displayed is mockup data

One design feature that has allowed the platform to facilitate increased information exchange and cooperation between communities and security providers is its SMS (“texting”) broadcasting system. Where previously, word-of-mouth was the only means of communication about security incidents in the region, now individual citizens can text an incident report to the Elva platform and designate that the SMS be broadcast from Elva to the community-at-large. Shortly thereafter, Elva personnel alert authorities to the security incident and then issue a community-wide SMS about the incident and what actions that authorities are taking to secure the situation.

*Note: Data displayed is mockup data

Due to this real-time community-wide security incident communication system, local communities are provide with an adequate amount of timely and reliable information about security incidents in their area and planned responses by security providers.

Other Uses of the Elva Platform

Elva was used to carry out pre- and post- election monitoring for Georgia’s 2012 elections. The platform is being used to map and analyze the alarming number of traffic fatalities that occur on Georgia’s streets and highways. Regional focus groups are using the Elva platform to collect and map data about Georgia’s solid waste infrastructure. Although Elva already has the support of UNDP and the East-West Management Institute/USAID, it is attracting an increasing amount of attention from core groups that operate within the international humanitarian community. With the upcoming launch of Elva 2.0 and its accompanying mobile application, it is highly likely that the Elva platform will be integrated into an increasing number of humanitarian and development projects across the globe.