DARA Big Data: 2019 Extension

Lead Research Organisation: University of Manchester
Department Name: Physics and Astronomy

Abstract

DARA Big Data is building research capacity around Big Data and the fourth industrial revolution in Southern Africa through high quality education and research. These high-value technical skills are applicable not only to scientific research but also to the space sector, as well as numerous other industrial and commercial sectors where diversification, technological upgrading and innovation are key drivers of economic growth. Moreover the key research areas of DARA Big Data are well-aligned both with the UN Global Goals and the African Union 2063 vision for advanced technologies, in order to improve welfare at the same time as targeting economic growth.

DARA Big Data provides bursaries for students from the partner countries of the African VLBI Network (AVN) - Botswana, Ghana, Kenya, Madagascar, Mauritius, Mozambique, Namibia and Zambia - to study for MSc(R) and PhD degrees at universities in South Africa and the UK. These degrees are in the three data intensive DARA Big Data focus areas of astrophysics (Astro Big Data), health data (Health Big Data) and sustainable agriculture (Agri Big Data).

In addition to providing studentship bursaries, DARA Big Data also works in partnership with South African SKA project (SKA-SA), now incorporated into the South African Radio Astronomy Observatory (SARAO), on the broader Big Data Africa program. Big Data Africa provides training workshops in machine learning, big data techniques and data intensive methodologies across the three DARA Big Data focus areas. These workshops and training courses take place in South Africa and other AVN countries, and are open to students from across the AVN country network who are currently in the honours year of their undergraduate degree or who are already pursuing a masters or PhD level research degree in those countries.

Planned Impact

This proposal requests an uplift to the existing Newton Fund DARA Big Data project, which is building research capacity around Big Data and the fourth industrial revolution in Southern Africa through high quality education and research. These high-value technical skills are applicable not only to scientific research but also to the space sector, as well as numerous other industrial and commercial sectors where diversification, technological upgrading and innovation are key drivers of economic growth. Moreover the key research areas of DARA Big Data are well-aligned both with the UN Global Goals and the African Union 2063 vision for advanced technologies, in order to improve welfare at the same time as targeting economic growth.

The primary legacy value of the DARA and DARA Big Data programs is the establishment of a network of skilled researchers who can underpin the future operation of the African VLBI Network and ensure scientific return for the African partner countries of the SKA as part of the stable data intensive research and education system required to enable data driven economic development for the fourth industrial revolution (Activity 1). Whilst this primary legacy creates a data intensive research community using radio astronomy as an existing source of big data in Africa, other aspects of the program act to increase the effectiveness and reach of that community. The multi-disciplinary nature of the DARA Big Data program ensures that the translational nature of data intensive skills is not neglected and that as Africa builds new databases addressing different priorities, including healthcare and agriculture - both ambitions under the African Union vision 2063, these skills can be propagated into those new areas.

The training in policy engagement (Activity 2 & Activity 3) provided through DARA Big Data is creating a cohort of young scientists who can effectively communicate with government about data-driven development. The legacy value of this training is the improved definition of policy in the 4IR going forward and the increased engagement of researchers with policy stakeholders.

Finally, to increase the reach of the program into the informal education arena, Activity 4 will provide resources that can be used beyond the program itself. These resources will be available to support the activities and growth of the tech sector beyond the tertiary education system, they will also promote engagement with STEM in Africa and beyond.

Publications

10 25 50

publication icon
Hosenie Zafiirah (2020) Imbalance Learning for Variable Star Classification in arXiv e-prints

publication icon
Ntwaetsile K (2021) Rapid sorting of radio galaxy morphology using Haralick features in Monthly Notices of the Royal Astronomical Society

publication icon
Bowles M (2021) Attention-gating for improved radio galaxy classification in Monthly Notices of the Royal Astronomical Society

publication icon
Vafaei Sadr A (2019) DeepSource : point source detection using deep learning in Monthly Notices of the Royal Astronomical Society

publication icon
Hosenie Z (2019) Comparing Multiclass, Binary, and Hierarchical Machine Learning Classification schemes for variable stars in Monthly Notices of the Royal Astronomical Society

publication icon
Smith R (2020) The Cloud Factory I: Generating resolved filamentary molecular clouds from galactic-scale forces in Monthly Notices of the Royal Astronomical Society

 
Description SARAO 
Organisation South African Radio Astronomy Observatory
Country South Africa 
Sector Public 
PI Contribution UoM leads the UK side of this Newton Fund project. The project provides training workshops in Southern Africa as well as bursaries for UK graduate study (PhD & MSc(R)) for students from Southern Africa. The project runs the Big Data Africa training program.
Collaborator Contribution SARAO drives the South African side of the project. The project provides training workshops in Southern Africa as well as bursaries for SA graduate study (PhD & MSc(R)) for students from Southern Africa. SARAO has delivered the very successful Data Science Intensive (DSI) programme in both 20/21 and 21/22 under the DARA Big Data banner, which has provided intensive skills training for about 40 students so far. In March 2022 DARA Big Data and SARAO co-hosted a 3 day virtual event which spanned the African continent; Africa Women in Data Science. This was attended by up to 160 people.
Impact This project is multi-disciplinary between astrophysics, health and sustainable agriculture research. Joint media statement on the outcomes of the 5th Ministerial Meeting of the Square Kilometre Array (SKA) African Partner Countries: "The Ministers noted the importance of the Big Data Africa project for both astronomy as well as more general preparations for the fourth industrial revolution. Given the significance of big data and cyber-infrastructure in economic development at both national and regional level, it was agreed that South Africa and Namibia would explore ways of integrating the Big Data Africa activities in Southern Africa into the SADC Industrialisation Strategy, taking into account prior work within SADC, and liaising with the other members of SADC. Kenya will explore a similar intervention in East Africa, and Ghana in West Africa. It was recommended that this work be presented to the African Union through the intervention of the Ministers."
Start Year 2018
 
Description DARA Big Data Hackathon - SGAC 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Postgraduate students
Results and Impact The DARA Big Data SGAC hackathon was organised in collaboration with the Space Generation Advisory Council (SGAC), an organisation of young space professionals and enthusiasts. It was held in November 2020 and was fully remote. The participants were from 12 different African countries and were grouped into international teams with others they didn't know before. Tutoring was all online.
Year(s) Of Engagement Activity 2020
 
Description DARA Big Data Hackathon - Zambia 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Postgraduate students
Results and Impact Two day hackathon event organised at University of Zambia. Hackathons are short, introductory-level data science training for students that usually take place over a period of two to three days. Participation of students in the hackathons gives them an introduction to data science and machine learning skills through a combination of tutorials and hands-on training using practical or real-life data sets.
Year(s) Of Engagement Activity 2020
 
Description Fanaroff Lecture 2020 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Undergraduate students
Results and Impact DARA Big Data ran the Fanaroff Lecture 2020, a lecture on science communication for policy engagement aimed at early career scientists
Year(s) Of Engagement Activity 2020
URL https://www.eventbrite.co.uk/e/fanaroff-lecture-2020-tickets-91803078479