The newly launched University of Johannesburg (UJ) Centre for Applied Data Science (CADS) aims to bring together computer and domain scientists to promote and accelerate data-intensive teaching with the goal of fostering research and education in data-driven knowledge discovery.

The UJ College of Business and Economics' School of Consumer Intelligence and Information Systems director Professor Mercy Mpinganjira notes that the centre, which is a collaboration with the Toulouse Business School Education, will focus on infusing science study by promoting various institution-wide and outreach activities.

“The centre is a multidisciplinary space in which different departments and disciplines are working together, including on agricultural research, on which many of our people on the continent rely for a living, as well as transport, supply chain management and marketing research. The aim is for the research to be impactful as it affects and must improve the lives of our people,” she said during a virtual launch of the centre.

There are ongoing research projects at the CADS in post-harvest waste, integrating transport options in the City of Johannesburg and beverage consumption, besides others, illustrated UJ CADS head Professor Hossana Twinomurinzi.

“Data is the bedrock of the Fourth Industrial Revolution, and is already changing how we work, live and communicate and is reshaping governments, education, healthcare, industries and commerce. The ability to use and analyse data will be one of the biggest drivers of businesses in the future,” said UJ vice-chancellor and principal Professor Tshilidzi Marwala.

“Against this backdrop, the centre aims to provide evidence-based insights to unearth new ways of working as industries, organisations, communities and societies,” he says.

APPLIED DATA SCIENCE

The October 7 virtual launch of the CADS formed part of UJ's Artificial Intelligence (AI) Day virtual conference. A panel of researchers involved with data science research on post-harvest agriculture, marketing, smart mobility and learning at the new centre provided insights into their research.

Professor Fawole Olaniyi highlighted that the application of data science in post-harvest agriculture is being used to investigate the complex value chain in which a lot of the commodities are lost.

“The world produces enough food to feed 1.5-times its population, but 850-million people go hungry every day. If we prevent food waste, we would be able to feed everyone and minimising losses would also provide an economic benefit, enable the industry to expand and ensure that nutrition and food are provided to the most vulnerable,” he said.

This can only be done by improving value chains. Agricultural value chains involve many stakeholders and generate lots of information, and the research at the CADS focuses on making sense of the information. The centre has a post-doctoral candidate who is doing research on applying machine learning for identification, classification and sorting of fresh commodities, he added.

Further, transport economist Professor Rose Luke said her team's work linked closely into what Olaniyi was working on, as they were working not only on transport in general, but also ensuring that the right products get to the right markets.

“Currently, we are looking specifically at mobility challenges, as the City of Johannesburg does not have adequate public transport and needs to look at ways to optimise what is currently present in the environment and use existing resources better,” she highlighted.

Congestion, accidents, disruptions owing to load-shedding, urban sprawl, long travel times and high costs of transport are common mobility challenges and solely adding more capacity through infrastructure runs into funding and other challenges, Luke noted.

Additionally, while there are plenty of different modes of transport, the system is fragmented and is not working as a whole. There are high levels of private car use. There is also some duplication in terms of funding and not the best use of resources on some subsidised corridors.

“We need to look at smart mobility solutions to leverage what is present and what we can do to resolve some of these challenges. To do this, we need readily available data.

“The idea is to ensure that everyone's individual mobility needs are looked after and to make smarter use of what we have to reduce congestion and develop faster, more sustainable transport solutions,” she stated.

Smart mobility looks at how costs can be reduced and how people can be moved in a sustainable or more sustainable way. The research questions based on this are generally about how smart mobility can be defined in the City of Johannesburg, and then what infrastructure, technologies and logistics are required for smart mobility in a city like Johannesburg, explained Dr Joash Mageto, who works alongside Luke at the centre.

The aim is to develop a holistic framework for mobility, the environmental, social and economic benefits, potential challenges and also how solutions would be adopted and accepted, he said.

“We found that government cannot participate in an effective way to provide transport as a service but should help create smart mobility solutions by providing the proper laws, regulations and infrastructure that can support a smart mobility ecosystem,” he noted.

Additionally, prospective students and industry practitioners at the CADS should have proper domain knowledge to serve as a basis on which to interpret the data, Mageto added in answer to a question on what prospective students and industry practitioners should expect during courses at the centre.

Meanwhile, Siyabonga Mhlongo was researching the use of data sciences to improve teaching and learning at the CADS.

There are greater volumes of data being generated, as lots of teaching and learning activities have moved online. The research is investigating how this data can be used to make the education journey more exciting and impactful for learning, teachers and administrators, he said.

“Learning analytics digs into this data to see what has happened and to draw insight on what is happening within a teaching and learning environment. This is a new field and we are only beginning to understand the value that data can bring to the teaching and learning environment.