Building fertile ground for data science in Uganda
Supporting locally-driven data projects: the AI Research Group experience
There is growing activity and interest in data science on the African continent, but the field is still nascent. For it to develop, it is important to create an ecosystem in which organizations working on different projects can cross-pollinate ideas and share their experiences.
Pulse Lab Kampala works with forward-thinking partners from private and public sector, academia, and research institutions to strengthen local and regional capacity for integrating real-time solutions into programmes to support sustainable development. One of the Lab’s long-standing collaborations is with the Artificial Intelligence Research Group (AI Research Group), a group specialized in the application of artificial intelligence and data science to problems in the developing world. The group operates out of the College of Computing and Information Sciences at Makerere University in Kampala.
AI Research is comprised of 26 individuals who are specialized in the application of artificial intelligence and data science including, for example, the use of machine learning, computer vision and predictive analytics. The group was founded in 2009 by Dr. John Quinn, who now leads the data science team at Pulse Lab Kampala.
Data science tools to advance agricultural development
Applications developed by the group have been presented during workshops organized with Pulse Lab Kampala and have sparked the interest of various stakeholders, including Government agencies and UN partners. Below are two examples of tools using mobile data to advance agricultural development in Uganda.
One application developed by AI Research is a mobile auction market that links farmers to traders, called Kudu. Any farmer or trader anywhere in the country can send a message in the system using any type of phone. The application computationally matches best compatible buyers and sellers and informs them of who they have been matched with. By matching farmers and traders with the use of algorithms, challenges such as the inefficient flow of information, or dependency on intermediaries can be overcome. The system encourages competition in the market and limits unnecessary movement of both farmers and traders.
Another application developed by the AI Research Group is Mcrops, a diagnostic tool for diagnosing viral crop diseases in cassava plants. The tool processes images of cassava plants taken with a camera or smartphone in cassava fields and scans for crop diseases. Sick plants are flagged in real time, which allows farmers to take action and stop the spread of the disease. The information can also be combined with spatial modelling to calculate the rate of the spread of the disease in real time.
Fostering collaboration and knowledge-sharing
Once a month, during the Lab’s data science seminars, experts from the AI Research Group share their latests application functionalities and brainstorm with participants. This is part of the Lab’s ongoing efforts to foster an ever growing network of passionate data scientists by reaching out to students and young people interested in the field.
Also, in collaboration with an even wider array of stakeholders including the AI Research Group, Pulse Lab Kampala hosted the second Data Science in Africa event in June of this year. Listen to presentations by the developers of Kudu and Mcrops directly from the event.
The fruitful collaboration with the AI Research Group and Makerere University has further contributed to building a robust academic community with other institutions on the continent such as the Stellenbosch University in South Africa and the Dedan Kimathi University of Technology in Kenya.
This is an exciting time in Africa and we hope that such interventions continue to resonate with the broader global community in order to advance data science and analytics in support of the Sustainable Development Goals.
To receive invitations for the monthly data science seminars organized at the Lab, email: firstname.lastname@example.org