Media monitor and chatbot for Tanzanian Red Cross
Location: Dar Es Salaam, Tanzania. Financed by the World Bank.
The variety of water related problems the Tanzanian population faces, has a profound impact on daily life. Among these problems, floods are a constant threat, occurring regularly and often result in fatalities. Especially in informal urban areas in Tanzania, that are part of Dar es Salaam, the impacts of floods are large, because informal settlements tend to be located in flood-prone areas, infrastructure and building quality is poor and maintenance of infrastructure and solid waste control is lacking.
To respond to the situation, disaster managers such as The Tanzania Red Cross Society (TRCS) need to know where new floods are happening, and form a solid situation awareness in short time. To stay up-to-date, they have been checking various online media data manually, and have been using WhatsApp to stay in touch with staff and volunteers. However, as the communication increases, it proves increasingly hard to keep track of the information manually, and contain a good overview to base their response on. As a result, their aid is less effective.
The Tanzanian Red Cross Society requested FloodTags to help them structure monitor floods from the media and organise the abundant direct messaging between themselves and the volunteers.
As a first measure, we started collecting all the Swahili news media in real-time. From the news, we isolate the media reports on floods and after filtering we present them on the Floodtags dashboard. This way, the TRCS can stay up-to-date without spending a lot of human resources to monitor it all. In addition, we developed a chatbot in Telegram to collect information more easily, connecting to their existing communication procedure.
With the software:
- TRCS volunteers and staff able to report on new floods easier and with richer content, via the Telegram chatbot. There are currently 129 disaster managers active in the group.
- TRCS central staff is able to anticipate more quickly and geographically more precise, based on more facts from the ground (media data as well as reports via direct messaging)
- TRCS management is able to submit a new DREF (Disaster Relief Emergency Fund) with much more evidence, and have a higher chance to receive additional funds for needed emergency aid.
TRCS has been using FloodTags in their operations since 2018. Plus they share the data with other stakeholders in disaster management in Tanzania, such as DarMaert (Dar Es Salaam Disaster Management Committee) and PMO (the Prime-Ministers Office).
The people of Tanzania share what they see, and the disaster managers listen and anticipate.
Messengers like WhatsApp and Telegram have found their way into water and disaster management, without any preset plan. Within only a few years, they are now widely used in the disaster management execution, considered an important source of information during floods. But there are some serious downsides: Information cannot be stored other than in the messenger, so they need to be manually copied elsewhere for any overview. Plus the information given is often incomplete or lacking details.
Additional to the media monitoring that Floodtags has practiced in many different settings, the improved use of user generated data such as from direct messaging is very promising. For this reason, FloodTags supports professionals benefit from their internal communication via messengers and help them engage via chatbots for more information from the field. The chatbot is optimized for smartphones, but it can also be used on SMS and USSD.