FloodTags monitors online media data in real-time, to detect new events and get immediate from-the-ground coverage. Effective for flood monitoring and response like at the Red Cross Operation Centre in Manilla.
Historic event detection
By scraping and analysing hundreds of thousands online articles, FloodTags detects and characterizes past events for hydro-meteorologic model calibration, validation, baseline studies, trend-analyses, ...
Floods & Disasters
We started with flood management and this is still the anchor of the company. Building on this experience, the software we applied on floods now forms the example for various other water management applications.
Who we are
FloodTags analyses online media and user generated content for water management and food security. Data sources include online news articles, blogs, forums, Twitter, Facebook public pages, and we connect to messengers such as WhatsApp and Telegram. We analyse these sources by using a mix of artificial intelligence, natural language processing and combinations with external data sources, including satellite imagery. Applications comprise real-time and historic (trend) analysis for emergency flood response, water scarcity management, water conflict resolution, integrated water management, food security and other water- and development related applications.
FloodTags believes that citizens and local communities are key resources in any social and water management problem. By listening to and anticipating on local perspectives, decision makers can take supported measures and responders have direct insight in the assistance needed on-the-ground.
Collect Online Media
Citizens and journalists describe new water crises in thousands of online messages. FloodTags collects their observations from Social Media, Blogs, Forums, Online News
Filter and Enrich
FloodTags and partners analyse the data using a mix of natural language processing and hydro-meteorologic enrichment, including remote sensed data
Share to Client
The Information is shared as a Service (IaaS) via a website and API that connects to client-side software. Applications include an array of water applications.
Collaborate with us
Floodtags has a partner network of top universities and institutions, with whom we improve the filters & enrichment algorithms for a wide range of use cases. Among our partners are Deltares, Radboud University Nijmegen, VU Amsterdam, Delft University of Technology and NASA. If you are a reseacher, anywhere in the world, working on social media for development: Feel welcome to contact us!
Publications with FloodTags
Geoparsing for Disaster Resp. (de Bruijn et al., 2018)
Probabilistic Flood Extent Est. (Brouwer et al., 2017)
Flood Key Term Inflections (Hurriyetoglu et al., 2016)
Action-based flood forecasting (Coughlan et al., 2016)
Near real-time flood maps (Eilander et al., 2016)
Early Flood Detection (Jongman et al., 2015)
Tweet Stream Analysis (Hurriyetoglu et al, 2015)
Just by looking at it from our desk, we can see the level of damage, the water levels, people that call out for help harvested from the different media considered in the tool
- Renatus Mkaruka, Head of Disaster Management at the Tanzania Red Cross Society -
End users feedback into the system. They can take ownership for the tool and shape its ongoing development to ensure it maintains context specificity in a changing environment
- Brigitte Rudram, Red Cross Climate Centre -
FloodTags is a big help for us, especially for the fast-tracking of
non-passable areas and damages
- Mitch, Operation Centre of the Philippine Red Cross in Manilla -
Tweets turned into flood maps that could help save lives
BARCELONA (Thomson Reuters Foundation) - Spontaneous tweets about major floods are being turned into a mapping tool that could be used by emergency services and disaster response teams to save lives and provide aid, Dutch researchers said. When a crisis strikes, people increasingly find out about it from social media, as individuals and groups take to the internet to spread the word.
After the Indonesian capital Jakarta was hit by floods this February, related tweets peaked at almost 900 a minute, with a significant number including information about location and water depth, according to a joint study by two Dutch organizations, Deltares and Floodtags. Read more