Research Articles (Computer Science)

Permanent URI for this collectionhttps://hdl.handle.net/10566/52

The research papers in this collection represent the work of several projects.

Browse

Now showing 1 - 20 of 185

GNNRec: a graph-based neural network model for efficient and robust product recommendation
(Institute of Electrical and Electronics Engineers Inc., 2026) Landry, Mbale Kasunzi Mbaherya; Merlec, Mpyana Mwamba; Maria, Kinyanta Wa Kinyanta
Product recommendation systems are fundamental to modern e-commerce platforms, yet existing approaches face significant challenges related to data sparsity, cold-start problems, and scalability limitations in large-scale applications. To address these issues, this paper introduces GNNRec, a graph-based neural network model for efficient and robust product recommendation. GNNRec is a comprehensive recommendation framework that extends the LightGCN architecture by incorporating multi-relational user–item interactions, heterogeneous node embeddings, and multi-task learning. The model integrates rich semantic features and edge-weighted graph propagation to capture complex user preferences and item characteristics effectively. GNNRec mitigates cold-start effects through graph-based propagation and content-aware initialization and addresses data sparsity via adaptive sampling strategies. Extensive experiments on large-scale real-world datasets demonstrate that GNNRec significantly outperforms state-of-the-art baselines, achieving an AUC of 0.916 and consistently superior results across ranking metrics. In particular, GNNRec attains NDCG@10 of 0.0428, corresponding to a 4.4% improvement over the strongest baseline (NGCF at 0.0410). Moreover, GNNRec exhibits strong robustness under data sparsity (99.988%), improved cold-start performance (16.3% gain for users with 5–10 interactions and viable recommendations for zero-interaction users), and temporal variations, while achieving 42.4% faster training and 33.3% lower memory usage, confirming its scalability and suitability for practical deployment.
Land-market drivers of peri-urban land transformation in Mpingu and Malingunde extension planning areas, Lilongwe District, Malawi
(Frontiers Media SA, 2026) Nyirenda, Clement; Nyengere, Jabulani; Tchuwa, Frank
Introduction – Peri-urban land use and land cover change is accelerating across Sub-Saharan Africa, reshaping agrarian landscapes where smallholder farming, settlement expansion, and common resources increasingly compete. This study investigated the drivers of land-use transitions in Mpingu and Malingunde Extension Planning Areas of Lilongwe District, Malawi, two rapidly transforming peri-urban agricultural landscapes. Methods – The study integrated spatial analysis with household-level econometric evidence. Land use and land cover maps for 2004, 2014, and 2024 were used to assess landscape transformation, while household survey data from 279 farming households captured land scarcity, fragmentation, tenure conditions, and market-related pressures. Land-use transitions were analysed using a multinomial logit model grounded in a land-market political economy framework. Results – The spatial analysis revealed substantial landscape transformation over the two-decade period, with natural woodland declining by approximately 66%, cropland expanding by approximately 459%, and built-up settlement areas increasing by approximately 127%. Household data showed severe land scarcity and fragmentation, with a mean farm size of 0.46 ha and more than 80% of households cultivating multiple spatially separated parcels. The multinomial logit model demonstrated strong explanatory power, with LR χ2 = 147.32, p < 0.001, and McFadden’s Pseudo R2 = 0.38. A one-unit increase in aggregate land demand pressure raised the probability of forest conversion to cropland by 18.4 percentage points, p < 0.01, while peri-urban land market intensity increased the probability of cropland conversion to settlement by 24.7 percentage points, p < 0.001. Competing land-use pressure significantly accelerated the conversion of communal grazing land to cropland, increasing transition probability by 16.3 percentage points, p < 0.01. Tenure security reduced the likelihood of settlement conversion by 11.2 percentage points, p < 0.01, indicating its moderating role in land-use change. Discussion – The findings show that peri-urban land transformation in Mpingu and Malingunde is driven primarily by land markets, spatial accessibility, and institutional conditions rather than demographic pressure alone.
Towards IoT-Fog-ML integration for temperature break detection and prediction in fresh produce cold chains: a systematic review and architectural framework
(Frontiers Media SA, 2026) Taguta, Jeremiah; Nyirenda, Clement Nthambazale; Nturambirwe, Jean Frederic Isingizwe
Introduction – Globally, 1.3 billion tons of food is lost or wasted each year, negatively impacting food security, the economy, and the climate. Fresh fruits and vegetables (FFVs), with their short shelf life and temperature sensitivity, are the most affected. This study systematically evaluates the integration of Machine Learning (ML), Adaptive Learning (AL), the Internet of Things (IoT), and Fog computing for temperature-break detection and prediction in FFVs supply chains. It critically evaluates their individual and combined capabilities, identifying compounding barriers that prevent genuine real-time integration of these technologies, while assessing their performance and operational readiness for real-time cold chain monitoring. Additionally, the role of fog computing in enabling efficient ML/AL deployment at the network edge is investigated for real-time applications in dynamic environments, with an implementation framework provided. Methods – Based on the PRISMA framework, searches of Scopus, Web of Science, IEEE Xplore, ACM Digital, supplemented by citation and reference chasing, produced 830 pre?deduplication records, identifying 14 relevant studies. Results – From the 12 analysed unique-dataset studies, 7 (58.3%) collected data using Basic Sensors, 3 with WSN (25%), and 2 with IoT (16.7%). Of the 14 ML studies, 5 (35.7%) detected temperature breaks, 5 (35.7%) predicted FFVs' temperature values, 3 (21.4%) predicted internal temperature (IT) values of a cold room or container, and 1 predicted IT values and time-to-temperature breaks. None of the studies predicted temperature breaks (event occurrence) or even their causes, while a few detected these breaks and their predefined causes. Four (28.6%) studies used IoT data, but none enabled live ML inference. None of the reviewed studies includes Fog or AL. Discussion – An integrated IoT-Fog-AL framework is thus proposed to address these gaps. These areas require focus to proactively reduce temperature breaks, thereby minimising food wastage and its associated effects, while also enhancing supply chain resilience and food security.
Dynamic evolving neuro-fuzzy inference system-based maximum power tracking controller for variable speed WECS
(Elsevier Ltd, 2026) Mabiala, Floyd; Nyirenda, Clement; Raji, Atanda
Reliable MPPT models are essential for maximizing energy conversion efficiency in wind energy conversion systems operating under highly dynamic atmospheric conditions. While most artificial intelligence (AI) controllers such as ANN and ANFIS provide robust control surfaces, their real-time deployment is frequently constrained by a reliance on static parameters, a lack of online adaptability, or severe computational burdens. To address these limitations, this paper presents an intelligent, self-organizing MPPT controller utilizing a DENFIS for a variable-speed PMSG wind turbine framework. By leveraging the Evolving Clustering Method, the proposed DENFIS-MPPT controller dynamically instantiates, updates, and discards localized fuzzy rules online, enabling continuous adaptive learning and rapid convergence without requiring extensive offline training or pre-calculated datasets. The computational and transient effectiveness of the proposed architecture is validated using 3 kW wind turbine model simulated within the MATLAB/Simulink environment. System performance is systematically evaluated under stochastic and gust wind profiles, with execution and event-driven characteristics benchmarked via the Simulink and Solver Profilers against ANN and ANFIS configurations. The empirical results demonstrate that DENFIS achieves superior maximum power extraction over a wide operational envelope. Detailed computational profiling reveals that DENFIS yields an optimized average execution cost of (24.92 (Formula presented) per iteration compared to 130.00 (Formula presented) for ANFIS, achieving an 80.8% reduction in structural computational overhead while incurring only a marginal execution premium over the rigid ANN baseline (14.11 (Formula presented) ). In terms of transient performance, DENFIS successfully stabilizes electrical variations, restricting output active power overshoot to a minimized 5.47%, thereby outperforming both ANFIS (9.00%) and the baseline ANN. These findings demonstrate that DENFIS provides an optimal architectural compromise between micro-level processing efficiency and macro-level adaptive intelligence, offering a highly viable, scalable control paradigm for next-generation real-time WECS applications.
Land use and land cover dynamics and its implications for agrifood systems among smallholder farmers in central Malawi
(Frontiers Media SA, 2026) Nyengere, Jabulani; Tchuwa, Frank; Tholo, Harineck Mayamiko; Njala, Allena Laura; Malalu, Lucius; Chisenga, Chikondi; Nalivata, Patson; Mwase, Weston; Kathewera, Msaiwale; Matewere, Brenda; Jamu, Lackson; Kanjira, Jones; Chabwera, Macdonald; Kachamba, Daud Jones; Jamali, Andrew; Musekiwa, Takudzwa; Nyirenda, Clement; Barker, Adele; Masuku, Precious; Mwangwela, Agness
Introduction – In central Malawi, the conversion of forests and grazing land into cropland and settlements is increasingly shaping the limits of smallholder production and household food security. This study assessed the effects of land use and land cover change on maize productivity, household dietary diversity, and food insecurity among smallholder farmers surrounding Lilongwe City. Methods – The study integrated multi-temporal Landsat imagery (2004, 2014, and 2024) with cross-sectional household survey data collected from 279 households in Mpingu and Malingunde Extension Planning Areas. Land use change was quantified using proportional changes in key land cover classes, alongside household-level exposure indicators capturing forest loss and grazing land loss in the surrounding landscape. Results – Cropland expanded by more than 450% between 2004 and 2024, while forest and grazing land declined by up to 70%, coinciding with a 61% reduction in average household farm size. Econometric analysis shows that forest loss exposure significantly reduces dietary diversity and increases the likelihood of moderate-to-severe food insecurity by 63% for each 10-percentage point increase. Grazing land loss is associated with livestock decline and reduced manure availability, contributing to lower maize yields and increasing soil fertility constraints. Dietary diversity emerged as the strongest protective factor, reducing the probability of food insecurity by 42% per additional food group consumed. Discussion – The findings demonstrate that peri-urban land conversion creates interacting land-livestock-soil-nutrition feedbacks that undermine agrifood resilience in land-scarce smallholder systems. Policy responses should integrate land governance, protection of forest and grazing resources, nutrition-sensitive agricultural diversification, and peri-urban planning to reduce the risk of persistent food insecurity in low-income agrarian economies.
Optimizing immersive services with parallel in-network rendering and deep RL
(Institute of Electrical and Electronics Engineers, 2026) Glitho, Roch H.; Gherari, Manel; Maia, Adyson
This paper addresses the challenge of delivering low-latency, scalable immersive experiences by exploiting a hybrid continuum of cloud, edge, and In-Network Computing (INC) resources. Indeed, delivering low-latency, scalable immersive experiences requires the transfer of a large amount of digital assets of different sizes, many of them consisting of large, static scene elements corresponding to service-specific and user-specific components. We argue in this paper that such elements could be separated within an in-network rendering farm while dynamically caching popular assets and synchronizing rapidly changing, user-centric data at INC, Edge or Cloud nodes. Still all theses need to be orchestrated efficiently. To efficiently orchestrate these heterogeneous resources, we formulate in this paper a multi-objective optimization problem - maximizing resource efficiency, minimizing end-to-end latency, and maximizing user request acceptance. This optimization problem is then solved via a deep reinforcement learning (DRL) framework that adaptively assigns functions across all layers in real time. The purpose of our proposed popularity-based replication and pre-caching is to further reduce latency for the most frequently accessed assets, while we offload lightweight rendering operations directly onto programmable switches to cut down on round-trip delays. Extensive simulations, benchmarked against multiple baselines, demonstrate that our approach consistently maintains sub-20ms end-to-end delays and achieves superior resource utilization efficiency under dynamic workloads. These results validate the potential of integrating INC into the Compute Continuum and use a DRL-driven orchestration, both together allowing to meet the stringent Quality of Service (QoS) and Quality of Experience (QoE) requirements of next-generation immersive applications.
Juro: a retrieval-augmented generation AI chatbot for enhancing legal information access in resource-constrained settings
(Springer Science and Business Media Deutschland GmbH, 2025) Ngandu, Bernard; Mbale, Landry; Bagula, Antoine
Access to legal information remains a significant challenge in resource-constrained settings where the digitization of legal systems is still in its early stages. To address this issue, we developed Juro, an AI-based chatbot architecture utilizing a Retrieval-Augmented Generation (RAG) framework. Leveraging a curated dataset of over 8,400 legal documents, Juro provides a user-friendly platform that simplifies complex legal language and ensures information reliability through a robust source citation mechanism. This paper demonstrates the applicability of adaptable AI-driven solutions in low-resource environments, offering a flexible chatbot architecture that can be tailored to various contexts where information accessibility remains a critical challenge. The legal system in the Democratic Republic of Congo (DRC) serves as a use case, illustrating the potential of Juro in addressing similar challenges across developing countries.
A novel hp-growth model for enhancing community policing in South Africa
(Springer, 2025) Macingwane, Apiwe; Isafiade, Omowunmi Elizabeth
South Africa (SA) grapples with a rising crime rate, which poses challenges to the safety and economic growth of the country. Despite the limited literature on pattern-based models in SA, frequent pattern-based models, particularly Frequent Pattern Growth (FP-Growth) and Hyper Structure Mining (Hmine), have demonstrated utility in various research fields. This paper introduces a novel model, Hybrid Pattern-Growth (HP-Growth), which combines the strengths of FP-Growth and Hmine. A comparative analysis of the South African crime statistics (Stats SA crime) dataset’s computational time complexity, scalability, and memory usage revealed that HP-Growth and Hmine outperform FP-Growth. This study establishes association rule thresholds and emphasizes the importance of selecting the most appropriate pattern-based model for generating crime patterns. According to the study, HP-Growth outperformed the other two models on sparse datasets, whereas Hmine excelled on dense datasets. FP-Growth uses more memory and has a greater time complexity than HP-Growth and Hmine. The most suitable model was then integrated into the developed crime support system. The research outcome has practical implications for law enforcement in aiding strategic resource allocation and proactive policing for crime reduction and control.
A multi-platform electronic travel aid integrating proxemic sensing for the visually impaired
(Multidisciplinary Digital Publishing Institute (MDPI), 2025) Naidoo, Nathan; Ghaziasgar, Mehrdad
Visual impairment (VI) affects over two billion people globally, with prevalence increasing due to preventable conditions. To address mobility and navigation challenges, this study presents a multi-platform, multi-sensor Electronic Travel Aid (ETA) integrating a combination of ultrasonic, LiDAR, and vision-based sensing across head-, torso-, and cane-mounted nodes. Grounded in orientation and mobility (OM) principles, the system delivers context-aware haptic and auditory feedback to enhance perception and independence for users with VI. The ETA employs a hardware–software co-design approach guided by proxemic theory, comprising three autonomous components—Glasses, Belt, and Cane nodes—each optimized for a distinct spatial zone while maintaining overlap for redundancy. Embedded ESP32 microcontrollers enable low-latency sensor fusion providing real-time multi-modal user feedback. Static and dynamic experiments using a custom-built motion rig evaluated detection accuracy and feedback latency under repeatable laboratory conditions. Results demonstrate millimetre-level accuracy and sub-30 ms proximity-to-feedback latency across all nodes. The Cane node’s dual LiDAR achieved a coefficient of variation at most 0.04%, while the Belt and Glasses nodes maintained mean detection errors below 1%. The validated tri-modal ETA architecture establishes a scalable, resilient framework for safe, real-time navigation—advancing sensory augmentation for individuals with VI.
Towards automated chicken monitoring: dataset and machine learning methods for visual, noninvasive reidentification
(Multidisciplinary Digital Publishing Institute (MDPI), 2025) Klauck, Ulrich; Kern, Daria; Schiele, Tobias
The chicken is the world’s most farmed animal. In this work, we introduce the Chicks4FreeID dataset, the first publicly available dataset focused on the reidentification of individual chickens. We begin by providing a comprehensive overview of the existing animal reidentification datasets. Next, we conduct closed-set reidentification experiments on the introduced dataset, using transformer-based feature extractors in combination with two different classifiers. We evaluate performance across domain transfer, supervised, and one-shot learning scenarios. The results demonstrate that transfer learning is particularly effective with limited data, and training from scratch is not necessarily advantageous even when sufficient data are available. Among the evaluated models, the vision transformer paired with a linear classifier achieves the highest performance, with a mean average precision of 97.0%, a top-1 accuracy of 95.1%, and a top-5 accuracy of 100.0%. Our evaluation suggests that the vision transformer architecture produces higher-quality embedding clusters than the Swin transformer architecture. All data and code are publicly shared under a CC BY 4.0 license.
AI versus tradition: shaping the future of higher education
(Emerald Publishing, 2025) Venter, Isabella Margarethe; Blignaut, Rénette Julia; Cranfield, Desireé Joy
Purpose: This research aims to investigate the use of conversational artificial intelligence (CAI) in academic practice through the lens of activity theory, which emphasises the mediation of human actions by tools within a social context. Additionally, it seeks to determine if and how the results of qualitative analysis differ when using traditional qualitative analysis software tools compared to using artificial intelligence tools. Design/methodology/approach: A pragmatic approach to the research design was used. The data collection phase included a survey, with open- and closed-ended questions and was distributed to academics in four countries (South Africa, Hungary, Lebanon and Wales). The data analysis phase included a mixed-methods approach integrating and interpreting both types of data to leverage the strengths of both qualitative and quantitative insights. Furthermore, traditional qualitative analysis methods and artificial intelligence tools were used for the analysis phase, allowing for a comprehensive understanding of the interactions between academics and these tools. Findings: Younger academics used CAI more for research than teaching, with academics from the science faculty using it more for teaching, and business management lecturers using it more for research. While viewed positively, concerns arose about ethics and educational alignment. This research shows how CAI supports qualitative analysis by saving time and suggesting new directions. Originality/value: Using an “activity theory” theoretical lens, with a pragmatic approach, the research explores how CAI tools impact academic practices. The study enriches theoretical discourse and offers practical recommendations for education.
Designing a universal electronic health record system: creating a reliable national health database for AI and its application in healthcare in South Africa
(South African Medical Association, 2025) Henney, Andre; Nodikida, Mzulungile
Digital technology and artificial intelligence (AI) are transforming industries worldwide, and healthcare is no exception. The application of AI in medicine dates back to the 1950s, with several pilot initiatives reportedly conducted in Africa during the early 1980s. AI refers to a computer system’s ability to perform tasks typically associated with human intelligence. These tasks include learning, decision-making, visual perception and speech recognition. Fig. 1 illustrates the different levels and categories of AI. To effectively leverage AI and emerging technologies in healthcare, AI models require access to training datasets in which outcome variables (e.g. disease onset) are clearly defined. Establishing a centralised technological infrastructure, specifically a universal electronic health record (UEHR) system, for all South Africans is therefore a critical first step. Such a platform would enable the collection and integration of high-quality health data, supporting national healthcare planning and ultimately improving health outcomes.
Effect of native language on learning to program
(Association for Computing Machinery, Inc, 2025) Henney, Andre; Russell, Seán; Alaofi, Suad
The dominance of the English language in computer science across programming languages, documentation, instruction, and scientific publication is well recognized. This situation contrasts with the actual distribution of spoken languages in the world, where only approximately 5% of the population of the world are native English speakers and a further 15% are non-native English speakers (NNES). For NNES students learning programming or computing in universities, the dominance of the English language can present a challenge. This challenge can manifest in multiple forms such as keywords, technical documentation, tutorials, or even descriptive terminology that may only exist in English. This situation impacts international students in English-speaking countries, those studying in regions where English is the medium of instruction, and it also affects students in non-English-speaking countries who do not study in English. In all of these cases, students may have to deal with computer terminologies or other documentation that have no direct translations in their language. This working group aims to systematically investigate the extent that the English language presents a barrier to non-native English speakers in computing education, specifically in introductory programming courses delivered entirely in English to populations of non-native English speakers. This multi-prong effort is based on the existing literature, instructor observations, student experiences, and popular introductory programming textbooks. The diverse nationalities and localities of the working group members prime our work for varied perspectives on the topic.
A conceptual framework for agility in sociotechnical contexts
(South African Institute of Computer Scientists and Information Technologists, 2024) Lillie, Theresa; Gerber, Aurona; Eybers, Sunet
Organisational agility is crucial for organisations to thrive in dynamic business environments. While the Information Systems (IS) discipline recognises the need for IS to support organisational agility, current IS research has not sufficiently explained how organisations achieve agility given their sociotechnical contexts. Some scholars and practitioners propose scaling agility-building approaches from small software development teams to the enterprise level, and others argue that agility is not a predetermined outcome of linear processes, but instead emerges from intricate organisational contexts. Previous research proposed a conceptual model that identified the structural components of agility in IS. However, this structural perspective does not address the dynamic aspects of agility. To address this gap, two systematic literature reviews (SLR) were conducted to develop a conceptual framework for agility in sociotechnical contexts, which is the contribution this research makes to the IS field. The first SLR investigated frameworks that enable organisational agility. Consequently, the Cynefin framework was adopted to explain the dynamics of contextualised decision-making and agility. The second SLR identified the influence of heuristics on decision-making and dynamic capabilities. The resulting framework integrates the structural and dynamic aspects of agility in IS and explains how heuristics could potentially be managed to improve sociotechnical agility.
HEALeB: healthcare 4.0 enabled by iot, ai, and blockchain
(Springer Science and Business Media Deutschland GmbH, 2025) Bagula, Antoine; Kyangwa, Henriette; Mbale, Landry
As the volume and sensitivity of medical data continue to grow, ensuring data security and effective analysis has become increasingly challenging. This paper presents the design and development of HEALeB, a comprehensive system that integrates blockchain technology and machine learning (ML) to address these challenges. Blockchain technology offers a decentralized, immutable ledger that ensures data integrity, privacy, and traceability, while ML algorithms enhance predictive analytics and decision-making capabilities. By combining these technologies, HEALeB aims to overcome critical issues such as data tampering, unauthorized access, and the underutilization of decision-support tools. The system’s design leverages blockchain to secure Electronic Health Records (EHRs) and employs ML to provide advanced analytics for personalized and precise healthcare solutions. This approach promises to improve diagnostic accuracy, optimize health management, and enhance overall healthcare quality. The paper provides a detailed exploration of HEALeB’s architecture, its implementation, and its performance evaluation, concluding with insights into the system’s impact and future research directions in the realm of Healthcare 4.0
A multifactor comparative assessment of augmented reality frameworks in diverse computing settings
(Institute of Electrical and Electronics Engineers, 2023) Maneli, Mfundo A.; Isafiade, Omowunmi E.
Research and development on different augmented reality (AR) frameworks have come a long way when it comes to image tracking, object tracking, plane tracking and light estimation. However, there might be trade-offs and varying results obtained from different AR frameworks, depending on the use cases, and this is critical for consideration during immersive application development. Besides the current literature effort, this research proposes a multifactor comparative analysis of two core AR frameworks, which aims to analyze and evaluate ARKit and ARCore in diverse computing settings. This research developed a structural application which evaluated three major test parameters across ten devices spanning ARKit and ARCore. The first parameter relates to evaluating AR measurements using four different distance criteria. The second parameter evaluated resource utilization, relating to the central processing unit (CPU) and random access memory (RAM), while the last parameter evaluated plane detection based on light estimation.
Clustered data muling in the internet of things in motion
(MDPI, 2019) Tuyishimire, Emmanuel; Bagula, Antoine; Ismail, Adiel
This paper considers a case where an Unmanned Aerial Vehicle (UAV) is used to monitor an area of interest. The UAV is assisted by a Sensor Network (SN), which is deployed in the area such as a smart city or smart village. The area being monitored has a reasonable size and hence may contain many sensors for efficient and accurate data collection. In this case, it would be expensive for one UAV to visit all the sensors; hence the need to partition the ground network into an optimum number of clusters with the objective of having the UAV visit only cluster heads (fewer sensors). In such a setting, the sensor readings (sensor data) would be sent to cluster heads where they are collected by the UAV upon its arrival. This paper proposes a clustering scheme that optimizes not only the sensor network energy usage, but also the energy used by the UAV to cover the area of interest. The computation of the number of optimal clusters in a dense and uniformly-distributed sensor network is proposed to complement the k-means clustering algorithm when used as a network engineering technique in hybrid UAV/terrestrial networks. Furthermore, for general networks, an efficient clustering model that caters for both orphan nodes and multi-layer optimization is proposed and analyzed through simulations using the city of Cape Town in South Africa as a smart city hybrid network engineering use-case.
Community healthcare mesh network engineering in white space frequencies
(Institute of Electrical and Electronics Engineers, 2019) Bagula, Antoine
The transition from analog to digital television has availed new spectrum called white space, which can be used to boost the capacity of wireless networks on an opportunistic basis. One sector in which there is a need to use white space frequencies is the healthcare sector
An economic feasibility model for sustainable 5G networks in rural dwellings of South Africa
(MDPI, 2022) Maluleke, Hloniphani; Bagula, Antoine; Ajayi, Olasupo
Numerous factors have shown Internet-based technology to be a key enabler in achieving the sustainable development goals (SDG), as well as narrowing the divide between the global north and south. For instance, smart farming, remote/online learning, and smart grids can be used to, respectively, address SDGs 1 and 2 (ending poverty and hunger), 3 (quality education), and 7 and 9 (energy and infrastructure development). Though such Internet-based solutions are commonplace in the global north, they are missing or sparsely available in global south countries. This is due to several factors including underdevelopment, which dissuades service providers from investing heavily in infrastructure for providing capable Internet solutions such as 5G networks in these regions. This paper presents a study conducted to evaluate the feasibility of deploying 5G networks in the rural dwellings of South Africa at affordable rates, which would then serve as a pre-cursor for deploying solutions to improve lives and achieve the SDGs.
Cyber security education is as essential as “the three R's”
(CELLPRESS, 2019) Venter, Isabella M.; Blignaut, Renette J.; Renaud, Karen; Venter, Anja
Smartphones have diffused rapidly across South African society and constitute the most dominant information and communication technologies in everyday use. That being so, it is important to ensure that all South Africans know how to secure their smart devices. Doing so requires a high level of security awareness and knowledge. As yet, there is no formal curriculum addressing cyber security in South African schools. Indeed, it seems to be left to universities to teach cyber security principles, and they currently only do this when students take computingrelated courses. The outcome of this approach is that only a very small percentage of South Africans, i.e. those who take computing courses at university, are made aware of cyber security risks and know how to take precautions. In this paper we found that, because this group is overwhelmingly male, this educational strategy disproportionately leaves young South African women vulnerable to cyber-attacks. We thus contend that cyber security ought to be taught as children learn the essential “3 Rs”—delivering requisite skills at University level does not adequately prepare young South Africans for a world where cyber security is an essential skill. Starting to provide awareness and knowledge at primary school, and embedding it across the curriculum would, in addition to ensuring that people have the skills when they need them, also remove the current gender imbalance in cyber security awareness.

Browse

Recent Submissions