Hackathon at Eduardo Mondlane University in Mozambique
Hackathon participants in their tutor groups outside Eduardo Mondlane University in Maputo; tutors Dionisio Candido on the left and Ramiro Saide on the right
DARA Big Data held their first ever event in Mozambique in March 2021, at the Eduardo Mondlane University (UEM) in Maputo. The event was a 3-day hackathon to allow local students and young professionals to improve and develop their Python skills. The hackathon was supported by Dr Claudio Paulo of the UEM Department of Physics, who has previously worked closely with the DARA Project.
The two tutors who led the event in Maputo and who worked incredibly hard to make it a success were Ramiro Saide and Dionisio Candido. Dionisio is a former BSc Meteorology student at UEM who has a strong interest in data science; he is hoping to study scientific applications of machine learning at PG level. Ramiro, a former UEM Physics student, is currently studying an Astronomy Masters in Mauritius with the DARA Project, searching for habitable southern exo-planets. You can read more about his research and his astronomy outreach work here. He also attended the Big Data Africa School in Cape Town in 2019, an intensive 10 day data science workshop. Ramiro and Dionisio received extensive training and support from Dr Nikhita Madhanpall, who puts together hackathons in conjunction with DARA Big Data and the Office of Astronomy for Development (OAD). The hack platform and associated support was provided by the Inter-University Institute for Data-Intensive Astronomy (IDIA).
The hackathon was initially planned for mid-February and had received a huge amount of interest from local students, with almost 150 applications received for only 24 places. Participation at the event unfortunately had to be strictly limited to ensure that social distancing could be maintained. However in early February Mozambique went into a lockdown and the University closed; the hackathon was postponed until local COVID19 rates improved significantly. After a month Maputo came out of lockdown and event planning resumed very quickly to put everything in place. The 24 places were confirmed and two laboratories at the University were booked to give students as much space as possible; face masks and hand sanitiser were also supplied. As Portuguese is the official language of Mozambique the tutorials were all translated from English by Ramiro and Dionisio.
The hackathon began in the morning of Monday March 15th with a remote presentation on data science from Prof Anna Scaife, followed by Dr Tawanda Chingozha of the OAD talking about using science for development. Following this Dr Ricardo Ogando of the National Observatory of Brazil gave a talk in Portuguese about the huge amount of data generated in astronomy and how to translate it using data science. In the afternoon the tutorials started, ably guided by Ramiro and Dionisio who tutored several teams each. Some students had little prior experience of Python and with the tutorials being quite intensive they continued throughout the second day of the event.
On Wednesday March 17th Mr Ornelle Nhaca, a data science analyst from Mozambique’s Standard Bank, came to campus to deliver a presentation on generating revenue using big data. Following this students began the hard work of their hackathon task! The dataset they were using contained Satellite Earth Observation (SEO) images of Mozambique following Cyclone Idai, which badly impacted the country in March 2019. The students were asked to monitor flooding in their own country resulting from the Cyclone, a task with direct relevance for them. They used the SEO images to try and forecast the areas that might suffer the worst flooding as well as study the progression of floods over a period of time; this type of analysis can lead to more lives being saved when used at times of natural disasters. Each team compiled their findings into a slideshow and gave a presentation at the end of the final day, showing how they used the data and the conclusions their analysis led to.
The judging panel consisted of Ramiro, Dionisio and Dr Ogando. Students were quizzed on their findings after each presentation, which were all of a high standard. There were prizes for the two teams which were judged to have used the data most efficiently, worked well together and gave the most thorough explanations of their analysis. Two individual prizes were also given for best presenter and best team leader. Students were very engaged throughout the hackathon; post-event feedback showed that they had not only enjoyed taking part but felt that they had really accelerated their programming and team working skills. Many of them felt that they would now like to explore data science as a career option and could use the skills they had learned to improve their current studies. One student said, ‘With this hackathon I realise the importance of big data knowledge in my day to day life; I gained a new objective which is to follow the area of computing and informatics in order to contribute to the growth of my country’. Another commented that ‘after the hackathon I feel like a different person, because I had the privilege to see what trend the technological world is following and with this I have the opportunity to not be left behind.’
Dr Madhanpall was impressed by how much progress students made in a short space of time; ‘I was so pleased to see the event run successfully in Portuguese, huge thanks to tutors Ramiro and Dionisio, as well as Dr Ogando, for making all of this possible. The participants were incredibly dedicated and were able to produce impressive results and draw insightful conclusions in response to the hackathon task, despite some having limited prior experience with Python’.
Hackathon tutors Ramiro and Dionisio were very proud of the students that they mentored through the hack and saw a lot of potential for them to develop further. Ramiro said that he found the workshop very inspiring; ‘I’m delighted to have been a part of this to transmit useful skills to students. I really enjoyed the student dynamics, they were all lovely people. We hope that the hackathon brought awareness to our universities and the general public of how big data and data science can solve real-world problems’. Dionisio noted that he was really pleased to have been involved in hosting the event, which was a great opportunity for him to build both data science and leadership skills; ‘After this I'm sure I can do a lot more to develop my skills in big data. We did inspire one of the students, she will use Python and machine learning to process data for her final course project. I hope we can do similar events in the future so that we can introduce big data to more people in Mozambique’.