Twitter becomes powerful tool in conflict resolution and natural disaster response

The social networking service is being used for more than just short musings – it is also tackling crime and aiding in crisis situations.

Haitians walk through collapsed buildings near the iron market in Port-au-Prince on January 31, 2010. AFP PHOTO / JEAN-PHILIPPE KSIAZEK

When someone loses a cow in the village of Lanet Umoja, Kenya, and Francis Kariuki hears about it, he normally sends a short message on Twitter so people know where to return the animal once found.

But Mr Kariuki, who is better known by more than 17,000 followers as @Chiefkariuki, has also utilised the social media tool to combat conflict.

After fielding a phone call where he was told invaders were pushing into a local teacher's home, Mr Kariuki reported the incident in a tweet. That attracted nearby villagers, who gathered outside of the teacher's home and watched as the attackers fled the scene, according to a report by the AP.

"I am using Twitter as a tool for community policing, neighbourhood watch and crime-reporting activities," Mr Kariuki's Twitter profile reads.

The use of social media and other online tools to track conflict and crisis situations is growing, observers say.

Some of the most dynamic examples include digital crisis maps, which popped up on the web immediately following natural disasters that struck Chile, Colombia, Pakistan and the Philippines. One appeared just after the earthquake that severely damaged parts of Haiti in 2010, helping crisis response workers pinpoint exactly where people were trapped.

Some volunteers from around the world translated Haitian Creole used in thousands of text messages and tweets sent by people who were on the ground and surrounded by crumbling buildings. Other volunteers pored over the translations to update the map so emergency crews could be deployed into the right areas.

Patrick Meier helped spearhead this project while working as director of crisis mapping at Ushahidi, a non-profit tech company. He recalls one tweet in particular: 18 children in an orphanage just outside of Port-au-Prince had run out of clean drinking water.

"It took us a few hours to verify the GPS coordinates," says Mr Meier.

"Then there was a follow-up, and we got confirmation water was delivered."

These days, Mr Meier is trying to speed up how quickly information is gathered and tacked on to a digital map, by taking out some of the manual labour that eats up precious time during disaster response situations.

As the director of social innovation at the Qatar Computing Research Institute, he is working on a project known as artificial intelligence for disaster response (AIDR).

AIDR is being programmed to comb through information shared on sites such as Twitter and identify eyewitness accounts that could help humanitarian organisations.

Currently, observations - such as a bridge having just collapsed, for example - can be verified with between 70 and 90 per cent accuracy, Mr Meier says. The goal is to eventually pull tidbits of information from dozens of social media channels, and in numerous languages without having to rely on individual volunteers to sift through everything manually.

"Hurricane Sandy generated over 20 million tweets during that one week," Mr Meier says.

"You can't process 20 million tweets. This is a major challenge the humanitarian community is totally unprepared to deal with."

The use of technology has not been limited to mechanisms for responding to natural disasters.

In 2010, groups such as the United Nations development programme and PeaceNet Kenya came together to establish a platform for peace known as Uwiano - a Swahili word that translates as "cohesion".

The efforts focused on preventing conflict in Kenya, particularly during run-ups to elections. One initiative included organising a system for getting up-to-date information on incidents of hate speeches, incitement and threats of violence.

The data would then be relayed to different security organisations and peace committees, which would hold mediation sessions or take appropriate action.

Certain organisations, including non-profits but also tech companies, have also pushed into this space in recent years with visual depictions of crises or crimes.

SumAll, a start-up based in New York City, normally spends its time analysing data for marketers and helping businesses grow by showing them information about product sales and corporate fans or followers.

But the venture has also spun off a non-profit organisation, the SumAll Foundation, which employees formed after they agreed to each set aside 10 per cent of their equity in the company to create it. The foundation aims to do social good by analysing data in an era when the economic opportunities that drive human trafficking, for instance, can be measured and analysed.

In February, the group released an infographic on the number of slaves in the world today - 27 million, up from 25 million in 1860 - but purposefully presented the data in a jarring format as a satirical fashion ad.

The SumAll Foundation plans to harness social media in future campaigns by "finding partners that embrace technology and are likely to believe in using more technology to really make a difference in their works," says Dane Atkinson, the chief executive of SumAll.

"We think there is likely a great way to directly leverage social [media] for our upcoming initiatives as the activity isn't illegal and the community plays a bigger part," he adds.

Google has also stepped into this space.

In a corporate blog post published in August, Scott Carpenter, the deputy director for Google Ideas, wrote about a new way the company tried to visualise the global arms trade. He notes 60 per cent of all violent deaths are due to small arms and light weapons, and that three quarters of the world's small arms can be found in the hands of civilians.

Google's Creative Lab, in collaboration with the Igarape Institute, a think tank, produced an interactive tool where people can zoom in or out of different countries around a digital globe and see historical ammunition trading patterns over the years. The tool, Mr Carpenter argues, provides policymakers with underexplored insight within the field of conflict prevention and resolution.

But even proponents of these kinds of technologies say certain tools could be more effective, especially when it comes to preventing problems developing in the first place. Some argue they do not always address the root causes that keep certain communities at risk - such as ethnic discrimination.

In other words, the technologies are not responding to the tensions before they escalate into violence, but trying to minimise violence if an incidence occurs, says Zahra Ismail, a programme officer at the Institute for Peace & Justice at the University of San Diego's School of Peace Studies.

Mr Meier also notes most online crisis maps have not adopted dynamic geo-fencing features, which would help humanitarians when analysing dots on a map.

Geofencing is a feature in a software program that uses the global positioning system or radio frequency identification to define geographical boundaries, according to WhatIs.com A geofence is a virtual barrier.

"Many geofencing applications incorporate Google Earth," Whatis says, "allowing administrators to define boundaries on top of a satellite view of a specific geographical area. Other applications define boundaries by longitude and latitude or through user-created and web-based maps," it adds.

Mr Meier says, for example, a map could geo-fence different sections of a refugee camp and trigger an SMS or email alert to the appropriate aid workers whenever there was a 10 per cent increase of new incomers or perhaps a shortage of food or clean drinking water.

"There's still a lot to be done with respect to visualisation," he says.