Dr Neema Mduma is addressing the issue of student drop-out in secondary schools in Tanzania, using a machine learning model. Deployed via a web-based application, this model allows teachers and parents to identify and provide support to students in a situation of academic fragility. In the medium term, she would like to use machine learning to help improve health care in countries affected by a shortage of medical staff.
My education journey started in 1996 at Bungo Primary School in Morogoro and then in 2003 I joined the Kilakala Secondary School in Morogoro, which is a special school for talented students, where I studied both ordinary level and advanced level. I went to Tumaini University in Iringa where I studied Information Technology and later I joined The Nelson Mandela African Institution of Science and Technology, where I obtained a PhD in Information and Communication Science and Engineering.
My greatest inspirations come from women who are doing well in science, particularly astronauts, neuro-surgeons, women who work in machine learning and artificial intelligence. There is one person who I consider my role model, a Professor at Stanford, called Fei-Fei Li, who is doing work on AI and machine learning.
It was great to see three women among the winners of this year’s Nobel Prizes in Chemistry and Physics. I have always been encouraged to study sciences and, honestly, I have not experienced any discrimination or glass ceilings, but this will serve as an inspiration and encouragement for all young girls and women out there and show that we can shine in science.
My PhD research focused on addressing student drop-out in secondary schools in Tanzania using a machine learning model, an application called BakiShule. It was developed purposely to help education stakeholders to intervene early to reduce drop-out rates. We looked at a number of factors and data sets, and there are six that stand out: age, gender, parent engagement, both with the school and also in terms of homework and reading, and income, measured through number of meals per day. We can quickly identify those students most-at-risk and as a result take remedial action.
The Covid-19 pandemic affected my research experiments since the University was closed and I had no access to the computer lab to run my models for almost three months. But thanks to technology, I was able to collaborate more closely with friends and colleagues in Uganda, Kenya and elsewhere. We actually managed to write a big proposal for a research project and raise funding. So it shows you also the power of collaboration and that will lead to greater sharing of information and knowledge.
Dr Neema Mduma was a L’Oreal-Unesco Women in Science 2020 Laureate.