Keeping in mind the pace at which COVID-19 is spreading and putting people’s lives at risk, several data science communities have come forward from different nations to find an effective solution of the pandemic. With more than 1,90,535 confirmed cases worldwide, communities have sprung into action and launched hackathons around the world for experts from AI and ML who are expected to develop new data text and data mining techniques that will shed light on a number of questions related to COVID-19. The hackathons are placed with the hope to create a scientific breakthrough that will either eradicate the virus or find an effective vaccine.
In this article, we will have a look at some of the top hackathons launched to tackle the COVID-19.
Curated by the Allen Institute for Artificial Intelligence, the Chan Zuckerberg Initiative, Georgetown University’s Center for Security and Emerging Technology, Microsoft Research, the National Library of Medicine — National Institutes of Health, and the Kaggle AI platform owned by Google, a publicly available and machine-readable dataset defines what CORD-19 is. It consists of more than 29,000 scholarly articles, with 13,000 full texts about COVID-19, SARS CoV-2 and other coronaviruses.
Hosted by Kaggle, the datasets can be accessed by more than four million data scientists who are expected to define 10 tasks based on scientific questions laid down by WHO and the National Academies of Sciences, Engineering, and Medicine’s Standing Committee on Emerging Infectious Diseases and 21st Century Health Threats. Some of the questions included in the tasks are as follows:
- What is known about transmission, incubation, and environmental stability?
- What do we know about virus genetics, origin, and evolution?
- What has been published about information sharing and inter-sectoral collaboration?
The tasks have been explained in detail on Kaggle and as per the guidelines, all the submissions must be made in a single notebook and should be made public before submission. Participants also have the liberty to choose other datasets, but those should be publicly available on Kaggle, Allen.ai or Semantic Scholar.
Forecasting Corona Break by MachineHack
As the virus outbreak is being considered as one of the biggest threats to mankind as of now, our team of MachineHack felt the need to find a way to predict when the virus will slow down or rather get worse. Due to this reason, MachineHack, along with other community members, is examining the coronavirus and its effects on various nations.
The machine hack is based on the data released by Johns Hopkins University Center for Systems Science and Engineering (JHU CSSE), which can be found below here. The .csv file consisted of confirmed COVID-19 cases across countries till 18 March 2020 for three target variables such as confirmed cases, recovered cases and death cases. Each participant is to submit one solution each day and a maximum of three for a day. The results will be scored calculating the cumulative scores of daily submission.
Hack the Crisis by Garage48 & Accelerate Estonia
Estonian startup Accelerate Estonia collaborated with Garage48 and launched an online hackathon within six hours that witnessed the participation of more than 1000 participants from 20 different countries in order to deal with the COVID-19.
The hackathon was able to generate 96 ideas and solutions and also formed 30 teams to further work on the outbreak. One of the most attention gained solutions was Zelos, a platform that enables vulnerable people to connect with volunteers via call centres and task dispatch app to prevent isolation. Another Estonian Velmio came up with the idea of a data-driven app named Corona-Tracker that retrieves data from symptom logging and wearable devices to deliver health insights for users to monitor their risk and recovery.
Hackathon by Wuhan2020
Organised by Wuhan community Wuhan2020, this hackathon was recognised and supported by IBM Developer along with acting as the technical supporter as well. The Wuhan2020 community is a group of 300 volunteers from different countries who are trying to facilitate information exchange between the resource provider and those who need help. The hackathon was organised with the aim in mind to develop an application for public welfare. It looked at novel designs, prototypes business or social models to create an impact on society in the fight against COVID-19.
The hackathon was organised with an online approach to design and develop on the data shared by the Wuhan2020 community. Some optional topic included annotated data for outbreaks and innovative application of relevant technologies.
AI for Mankind
AI for Mankind was a COVID-19 hackathon on GitHub with the goal to come up with ideas and build applications to help communities in dealing with COVID-19. Some of the ideas that were shared are real-time temperature monitoring proposed by Wei Shung Chung, COVID-19 cases data exploration colab notebook proposed by Jared Yu and tracking of those people who may have come in contact with infected ones, proposed by Anna T and Jean-Phillipe Monfret.
The participants were asked to create a public GitHub repository for entry with MIT licence and were denied entry in projects containing confidential information.
With the disease spreading at a rapid pace, it can be cited that more hackathons should be organised by communities and countries to find a quick solution in terms of slowing down the spreading rate or to find a cure.