A hackathon is a fun, fast-paced, strenuous, yet highly rewarding event where developers/coders, designers, strategists, and enthusiasts come together to work on project ideas and build solutions. Hackathons are known for late-night coding in groups, with shared pizzas and screens – but not in these times of pandemic. This time, we do it remotely.
This hackathon is co-hosted by the University of Zambia (UNZA), DARA Big Data, the Office of Astronomy for Development (IAU-OAD) and the Inter-University Institute for Data Intensive Astronomy (IDIA). We have hosted several hackathons across Africa before, but this is the first one to be hosted remotely.
Usually, the participants learn to hone their big data skills using the IDIA/ilifu research cloud. That doesn’t change and demonstrates how cloud computing is independent to the location of researchers using it. This time, however, we are not all gathering at the University of Zambia. While the 24 student participants will be there, with two tutors, others will join in from places such as Spain, South Africa or the UK. We are using a combination of Zoom and Slack as a communications platforms, GitHub, and of course the IDIA/ilifu research cloud.
The theme of this hackathon is Natural Language Processing (NLP) and Machine Learning (ML). Students will collect and prepare COVID 19-related data directly from Twitter using their Application Programming Interface (API), which allows automated machine-to-machine between Twitter and software running elsewhere. They will then build a machine learning model to perform sentiment analysis on tweets from around the world (but in English) on COVID-19 topics.
The challenge in this hackathon requires basic knowledge of the Python programming language, which the participants have acquired experience with during this study year. The challenge for the organisers is to have a smoothly functioning mixed physical and virtual event. Good luck to all!