When the COVID-19 global pandemic started in March 2020, the Inter-American Development Bank sought to understand people’s needs quickly in order to inform their ongoing and varied communication, emergency response, policy and investment decisions. They made the aggregated data publicly available at bidciviclytics.citibeats.com in order to provide a public resource that governments, companies, organizations and startups across the region could use to generate solutions to the crisis.
To respond during the crisis, it was essential to have a system put into action quickly, as well as be able to understand the diversity of languages and local references used to explain the varied social needs resulting from the pandemic.
Citibeats’ social data platform, based on proprietary natural language processing technology and machine learning algorithms, makes it possible to unlock the analysis of big data about human behavior and turn it into actionable insights - regardless of the language, data source, or structure of the text. The technology in some cases can yield results more than 30 days earlier than traditional methods - like media, polls or surveys - which can help organizations manage emerging risks and detect early warnings on critical issues.
“We rely on data from Twitter’s enterprise APIs to give us the full picture of what is going on, so we can surface the most actionable information,” said Harry Wilson, Chief Product Officer at Citibeats. “Thanks to the powerful endpoints available with Twitter enterprise data access, we’re able to make more specific queries, meaning more specific data that can be acted on.”