UPSpace Repository

Faculty Research Collections

UP Research Output Collections

Featured

Recent Submissions

Evaluating a human-machine student intervention framework in higher education from legacy data
Combrink , Herkulaas Morkel van Eyssen (University of Pretoria, 2026)
This study aims to answer the question: “Can education be abstracted in a framework and conceptually studied for a student intervention process?” To do so, it designs and tests a student intervention framework that uses a systems approach and complexity theory to learn from different student contexts and recommend and apply systemic interventions for students. The framework is based on the assumption that education is a complex system that involves multiple interacting elements, such as students, teachers, curriculum, policies, and resources. The specific interventions and their impact on student success are not assessed in this work, as they depend on the expertise in the education domain. Rather, this work focuses on the framework that can be used to apply and evaluate different contexts and interventions. The study applies complexity theory and systemic intervention theory to understand the lens and the methods for studying the system. It also explains the education system in South Africa as the context of the study. One of the main challenges of the study is data sharing and data handling. To address this challenge, the study generates synthetic data from openly available tabular data and evaluates its conditional interdependence using different machine learning classification tasks. Then, it applies the same methods to a real-world education dataset from the University of the Free State. The study contextualizes the student intervention framework as a multi-armed bandit (MAB) stateless reinforcement learning problem and tests its performance and viability using probabilistic models. The results show that the probabilistic models yield the best results with the minimum required fine-tuning, and that they scale well to the real-world dataset. The results also indicate that the framework is viable for student intervention recommendations within the context of contextual bandits. The study concludes with a discussion of the implications and limitations of the framework, and suggests areas for future research.
Understanding black women opportunity entrepreneurs' networking behavior : navigating intersectionality in South Africa's entrepreneurial ecosystem
Pettersen, Inger Beate; Scheepers, Caren Brenda; Botha, Melodi; Moos, Menisha; Moos, Menisha (Emerald, 2026)
PURPOSE : Globally, black women entrepreneurs represent the fastest-growing group of entrepreneurs, driving significant economic and social impact. This research timely responds to calls aiming for varied research contexts and for investigating women entrepreneurs in non-Western contexts. This study examines networking within South Africa's entrepreneurial ecosystems (EEs) to contextualize theory across diverse cultural and socio-economic settings, thereby deepening understanding of the compounded barriers faced by black women entrepreneurs at the intersection of race and gender. DESIGN/METHODOLOGY/APPROACH : A qualitative research design was adopted to explore individual experiences while identifying patterns and connections that can enhance our understanding of black women's networking behavior to navigate the EEs in South Africa. FINDINGS : The study's findings reveal three unique themes which contribute to women's proactive networking behavior, involving cross-gender and cross-race relations. Furthermore, we learn how black women entrepreneurs exploit strategic networking opportunities to grow their tech ventures and navigate intersectionality through resilience of being part of a minority group. RESEARCH LIMITATIONS/IMPLICATIONS : EE actors should deliberately introduce allyship programs to enhance black women's position and belonging in various settings through, e.g. mentorship and training programs. Policymakers in other countries could gain key insights from the South African Broad-Based Black Economic Empowerment (BBBEE) policy, which positively influence women entrepreneurs' position in EEs. ORIGINALITY/VALUE : Our study links the Intersectionality Theory with Social Network Theory. Previous studies considered gender and networking behavior, without referring to the intersection between gender and race and how this intersection might influence networking behavior.
Affordable loss heuristic scale development and validation
Lew, Charlene; Smith, Sean Michael (Emerald, 2026-12)
PURPOSE : The study aims to develop a distinct and parsimonious scale of affordable loss. The affordable loss heuristic explains a way of reasoning in entrepreneurial investment decision-making in uncertainty. DESIGN/METHODOLOGY/APPROACH : The scale development process spans several phases including conceptualization, qualitative interviews and expert item review in the item development phase. In the exploratory phase, two studies are used to refine the instrument. In the validation phase, the scale is validated through confirmatory factor analysis and assessment of the invariance thereof. FINDINGS : The study produces a two-dimensional reflective scale of affordable loss, including dimensions of a loss heuristic in uncertainty and experimental mindset. The scale is validated and invariant across different geographical contexts. RESEARCH LIMITATIONS/IMPLICATIONS : The scale builds on the theory of affordable loss, presenting clearer insights into the nature and dimensions of this heuristic. It offers an empirical contribution by enabling the measurement of affordable loss as a construct in entrepreneurial decision-making. PRACTICAL IMPLICATIONS : An understanding of affordable loss gives practitioners the means to assess and develop entrepreneurial reasoning. ORIGINALITY/VALUE : The study presents a novel measure of the affordable loss investment heuristic, paving the way for future studies into its antecedents and outcomes.
Enhancing survival of Bifidobacterium spp. in yoghurt through oxidative stress adaptation
Thomashoff, Ursula Louise; Sibanda, Thulani; Buys, E.M. (Elna Maria) (Wiley, 2026-01)
BACKGROUND : Bifidobacterium spp. are widely recognised probiotic bacteria with well-documented health benefits associated with their incorporation into foods. Despite these benefits, maintaining their viability in oxygen-rich environments like yoghurt poses substantial challenges. AIM : This study examined the impact of oxidative stress adaptation on the viability, fermentation characteristics and physicochemical properties of Bifidobacterium spp. during yoghurt production and shelf-life at 4°C. METHODS : Oxidative (H2O2) stress-adapted and unadapted strains of Bifidobacterium bifidum, B. breve and B. animalis subsp. animalis, 6–7 log cfu/mL, were incorporated during yoghurt fermentation. Two independent yoghurt batches were produced for each treatment combination, with two replicates analysed per batch. Viability (via plate counts and PMAxx-qPCR) and physicochemical properties were monitored over 28 days at 4°C. MAJOR FINDINGS : Stress adaptation enhanced B. bifidum survival during fermentation by 1.0 log cfu/g compared with unadapted strains (P < 0.0001), with B. breve showing similar improvement. Although adaptation initially improved survival, this advantage diminished throughout shelf-life in both species. B. animalis remained stable. The plate count method showed a decline of B. bifidum and B. breve populations, while the PMAxx-qPCR method detected a significantly higher level of viable cells (P < 0.05) in the yoghurt. SCIENTIFIC IMPLICATIONS : Oxidative stress adaptation may enhance Bifidobacterium spp. viability and the functional value of probiotic yoghurt.
Numerical analysis of stabilisation for random hyperbolic systems of conservation laws
Chu, Shaoshuai; Herty, Michael; Kurganov, Alexander (Taylor and Francis, 2026)
This paper extends the deterministic Lyapunov-based stabilisation framework to random hyperbolic systems of conservation laws, where uncertainties arise in boundary controls and initial data. Building on the finite-volume discretisation method from [M. Banda and M. Herty, Math. Control Relat. Fields, 3 (2013), pp. 121–142], we introduce a stochastic discrete Lyapunov function to prove the exponential decay of numerical solutions for systems with random perturbations. For linear systems, we derive explicit decay rates, which depend on boundary control parameters, grid resolutions, and the statistical properties of the random inputs. Theoretical decay rates are verified through numerical examples, including boundary stabilisation of the linear wave equations and linearised shallow-water flows with random perturbations. We also present the decay rates for a nonlinear example and for the linearised Saint-Venant system with source terms.