A team from UAE, Pakistan, and Saudi Arabia answer the call with AI
With countless COVID-19 related sources of information, a research assistant has never been more valuable
The rise of the COVID-19 pandemic resulted in a dire need for research efforts in the medical field during these times to come up with solutions to understand COVID-19, how to treat patients, and develop vaccines.
Due to the rapid acceleration in new coronavirus literature, it has become difficult for the medical research community to keep up. And in response to the pandemic, the White House and a coalition of leading research groups have prepared the COVID-19 Open Research Dataset (CORD-19) which includes over 200,000 scholarly articles, including over 93,000 with full text, about COVID-19, SARS-CoV-2, and related coronavirus information.
Therefore, in support of the 2020 Call for Code Global Challenge, my team, Huzaifah Saleem, Mofaq Althiyabi, Saad Tariq, and I built a solution, a Research Assistant that aims to help the medical community in answering high priority scientific questions using AI Technologies.
Among the 200,000 articles, we have used 9,120 scholarly articles from “comm_use_subset” and fed it to Watson Discovery to build queries of the most common scientific questions around coronavirus, viruses, symptoms, vaccines, etc. and trained our model to find relevant answers from the scholarly articles.
Then, we built an assistant using Watson Assistant and connected it to our Watson Discovery to show the top results of the queries.
Check out the GitHub repo for the project here, and dataset here. You can also learn more about the research assistant here.
Ready to take action? You could be a part of a Call for Code solution and make a real impact with your code. Contribute to an open source project today!
If you liked the story, be sure to give it a clap and follow Call for Code Daily for more tech-for-good stories!
