Artificial Intelligence for Digital Response

Artificial Intelligence for Digital Response (AIDR) is a free and open source platform to filter and classify social media messages related to emergencies, disasters, and humanitarian crises.^[1][2][3] It has been developed by the Qatar Computing Research Institute and awarded the Grand Prize for the 2015 Open Source Software World Challenge.^[4][5][6]

Muhammad Imran stated that he and his team "have developed novel computational techniques and technologies, which can help gain insightful and actionable information from online sources to enable rapid decision-making" - according to him the system "combines human intelligence with machine learning techniques, to solve many real-world challenges during mass emergencies and health issues".^[2]

It can be used by logging in with ones Twitter credentials and by collecting tweets by specifying keywords or hashtags, like #ChileEarthquake, and possibly a geographical region as well.^[7]

• It has been deployed in conjunction with UNICEF in Zambia to classify short messages related to AIDS/HIV received through the U-Report platform.^[8][9]

• AIDR was used for the first time during the 2010 Pakistan floods.^[10] The first real test of AIDR took place during the 2014 Iquique earthquake in Chile.^[7]

• Muhammad Imran delivered a keynote talk on the science behind the AIDR system at the International Conference on Information Systems for Crisis Response And Management (ISCRAM).^[11]
• Abdelkader Lattab and Ji Lucas also presented the system at the 2016 QCRI-IBM Data Science Connect event.^[12]

• Crowdmapping
• Digital humanitarianism
• Social media analytics
• Social media mining

• Official website
• AIDR on GitHub