39624 research outputs found
Sort by
Navigating the Paradox Materialism Sustainable Practices and WellBeing in Emerging Markets
No full textCross-domain few-shot classification for remote sensing imagery
No full textA thesis submitted in fulfilment of the requirements for the degree of Masters of Science, to the Faculty of Science, School of Computer Science & Applied Mathematics, University of the Witwatersrand, Johannesburg, 2025Deep learning has proven highly effective for scene classification tasks when substantial quantities of labelled data are accessible. However, performance decreases when applied to domains such as remote sensing which typically possess a limited quantity of labelled data across available datasets. Few-shot learning has been developed as one of the promising solutions to this problem. It has the ability to recognise new categories with minimal labelled examples, but it assumes that the training and testing data will exhibit identical feature distributions. This assumption is unrealistic in real-world contexts where data can originate from different domains and poses a challenge when a significant domain shift exists between the training and testing data. This dissertation aims to address these limitations by proposing the Cross-Domain Attention Network (CDAN). It is a network designed specifically to solve the issues that arise when there is a limited quantity of labelled data available and a significant domain shift exists between the training and testing data. The network proposed consists of a prototypical network as the base and three additions that contribute to the accurate scene classification of remote sensing imagery. Firstly, a cross-domain data augmentation technique is proposed with few-shot learning to reduce domain shift. The cross-domain data augmentation technique facilitates enhanced knowledge transfer between domains and increases the adaptation ability of the network, whereas few-shot learning reduces the network’s reliance on large labelled datasets. Secondly, a dynamic and focused attention module is proposed to improve discriminative capacity of the network by increasing the focus on important channels and spatial regions within images during training. Thirdly, an adaptive task aware loss is proposed to further enhance the network’s adaptive capacity by leveraging information in few-shot training tasks. Extensive experiments are carried out with different remote imaging classification datasets (RSICB, AID and NWPU-RESISC45) to prove that the proposed network alleviates concerns in a cross-domain few-shot (CDFS) classification setting.MMM202
Social Support Provisions for the Health Care Financing of Older People: An Exploratory Sequential Mixed-Methods Analysis
No full textLiterary reading as a socially responsive practice Implications for literature pedagogy at higher education
No full textThe Impact of liquidity on Bank profitability – Evidence from Zimbabwe and South Africa Commercial Banks
No full textA research report submitted in fulfillment of the requirements for the Master of Management in Finance and Investments degree, in the Faculty of Commerce Law and Management, Wits Business School, University of the Witwatersrand, Johannesburg, 2025This study examines the impact of liquidity and macroeconomic factors on bank profitability in South Africa and Zimbabwe, focusing on the distinct economic and regulatory environments in these countries. Liquidity is a vital component of banking, ensuring banks meet short-term obligations while maintaining financial stability. The study examines the relationship between liquidity and profitability, considering both internal bank-specific factors and broader macroeconomic conditions. Global research on this relationship presents mixed findings, with some studies showing a positive correlation, while others suggest an insignificant or negative effect. This research contributes empirical evidence from these two Southern African countries—Zimbabwe, marked by economic instability and hyperinflation, and South Africa, characterized by a more stable banking system. The study uses secondary data from 12 commercial banks in both countries, covering the period from 2012 to 2022. It applies fixed-effects regression models to analyse data on bank liquidity, size, asset quality, cost management, market structure, economic growth and inflation. The findings indicate that liquidity does not significantly influence profitability in either country, likely due to regulatory requirements ensuring adequate liquidity levels across banks. In contrast, the study reveals a negative relationship between economic growth and profitability, suggesting that during economic expansions, banks lower lending rates to capture market share, reducing profit margins. Inflation was identified as having no significant impact on profitability, possibly due to effective inflation risk management strategies employed by the banks. The study concludes that banks in both countries should focus on improving operational efficiency, capital adequacy, and cost management rather than prioritizing liquidity levels for profit maximization. It also recommends that banks adjust their business strategies in response to economic growth cycles to mitigate associated risks. Despite inflation's minimal impact, the research advises banks to maintain strong inflation risk management practices. The study provides valuable insights for policymakers, bank managers, and researchers looking to understand the dynamics of bank profitability in emerging markets.MM202
Impact of the Akwenda Intervention Program for cerebral palsy on caregiverperceived burden stress and psychosocial functioning A clusterrandomized trial in Uganda
No full text
Loyalty program design, retailer and country characteristics effect on retailer performance in Lesotho
No full textA research report submitted in fulfillment of the requirements for the Master of Business Administration, in the Faculty of Commerce Law and Management, Wits Business School, University of the Witwatersrand, Johannesburg, 2025Grocery retailers operate in a highly competitive environment with little differentiation between the products sold. Grocery retailers have resorted to loyalty programmes to attract and retain customers to maintain a share of the market. The present study investigated how the design elements of a loyalty programme in the form of rewards timing (immediate and delayed rewards), rewards type (direct and indirect rewards), and plural rewards affect the sales of a grocery retailer. The study also investigated how the retailer characteristics represented by the retailer pricing strategy and the country characteristics represented by competing loyalty programmes affect the sales of a grocery retailer. The study used a questionnaire as the research instrument. Two different questionnaires were used, one to solicit responses from consumers and the other sent to two store managers of Pick n Pay stores in Lesotho to get the perspective of the grocery retailers. The Statistical Package for Social Sciences was used for data analysis and hypothesis testing. Four of the seven hypotheses of this study were supported. A positive and significant relationship was found to exist between immediate rewards, indirect rewards, plural rewards and retailer sales. The strongest relationship was found between plural rewards and retailer sales, which indicated that consumers prefer plural rewards which combine both immediate and delayed rewards and direct and indirect rewards. A significant, negative relationship was obtained between competing loyalty programmes and retailer sales, which indicates that an increase in the number of competing loyalty programmes leads to a reduction in sales of the retailer. Hypotheses which were not supported and found to be not statistically significant were the relationships between delayed rewards, direct rewards, retailer pricing strategy and retailer sales. The study revealed that consumers in Lesotho prefer plural rewards over singular rewards. Thus, to reap maximum benefit from loyalty programmes in Lesotho, grocery retailers must have plural reward programmes in place. Those with existing loyalty programmes, like Pick n Pay, must update their programmes and those thinking of introducing loyalty programmes must think of plural reward programmes.MM202