top of page

Kushatha Ntwaetsile


PhD Astrophysics

University of Hertfordshire


Could you tell us about your background? 

I have a BSc (hons) in Computer Systems Engineering from the University of Sunderland, taken through the Botswana Accountancy College. I also have an MSc in Computer Science from Botswana International University of Science and Technology, where I also worked as a Graduate Teaching Assistant during taking my masters. Prior to starting my PhD, I worked as a Lab Demonstrator at the University of Botswana in the Computer Science department.


Please tell us about your PhD research.

My PhD is titled Machine Learning Classifications for Astronomical Radio Surveys. As astronomy is becoming more data driven, new tools such as machine learning will be essential. Algorithms can recognize patterns and details in large data sets, in a way that cannot be done with the human brain. It will be important to efficiently analyse this data in real time if possible. So, machine learning algorithms are the necessary tools for knowledge discovery that can help to recognise hidden patterns and understand relationships in astro data. My research uses unsupervised machine learning techniques for the automatic identification and classification of different transient phenomena in radio surveys. Currently I am working on the use of `Haralick' features for the automated classification of radio galaxies. Using simulated data from the Square Kilometre Array data challenge showing a large and deep extragalactic survey field, we demonstrate how the Haralick features of radio sources can be clustered automatically into morphological classes in an unsupervised manner. This provides a methodology to quickly label new and unseen galaxies and identify morphological outliers.

What do you plan to do when you have finished your studies?

Upon completion of my studies I want to continue with research as a postdoc for a couple more years and from there I would like to apply my data science and machine learning skills to the financial sector. I would love to work as a data analyst in banks. I also currently teach on a part-time basis as an online tutor so I would like to continue doing that as well.












What started your interest in astronomy and data science? 

My interest in astronomy started during the DARA basic training I took in Namibia in 2017; I was among the first cohort from Botswana. That is where I got to learn more about astrophysics and how I can incorporate my computer science skills into astrophysics. During my MSc I used supervised machine learning techniques to predict the stock market using social media data, I guess that is where my interest in data science really began.


What would your dream job be, and where?

Like I mentioned before I have an interest in big data analytics in banking. I would like to build algorithms that help in the detection of fraud and analysing transactions, so my dream job would be a data analyst in one of the big banks like Barclays or Santander.

What accomplishment are you most proud of so far?

Being awarded the Big Data scholarship to study for my PhD in the UK is the biggest achievement for me so far and I will forever be grateful to DARA Big Data for giving me this opportunity.

What advantages do you think there are for students with machine learning skills, particularly in Africa?

Because of new computing technologies machine learning has gained a fresh momentum and it can now be applied across various fields. With growing volumes of available data and powerful computational power they now have opportunities to build precise models that can analyse complex data and deliver accurate results.


What have you enjoyed most about the UK while you've been studying here?

The UK is multi-cultural, so I have really enjoyed meeting people from all walks of life, interacting with them and learning about their cultures and beliefs. Most importantly I have enjoyed travelling around the UK and just seeing different places and meeting lovely people.



Kushatha talks through her research at the 2019 Big Data Africa School, held in Cape Town

bottom of page