4. Electronic Theses and Dissertations (ETDs) - Faculties submissions

Permanent URI for this communityhttps://hdl.handle.net/10539/37773

Browse

Subject: search.filters.subject.Artificial Intelligence (AI)

Search Results

Now showing 1 - 5 of 5
  • Thumbnail Image
    Item
    The impact of artificial intelligence capabilities on organisational performance: an empirical study in the South African retail context
    (University of the Witwatersrand, Johannesburg, 2024) Cronjé, Dylan Christo
    Developing the ability to undergo digital transformation with Artificial Intelligence (AI) is increasingly crucial for retail organisations, given the rising occurrence of AI-driven activities within their organisations. This underscores the need to understand how retail organisations should structure themselves to leverage AI effectively and in what ways value can be attained. Within this context, this thesis investigates how AI capabilities can enhance organisational performance by prompting changes in critical organisational activities. Through a survey-based research approach, data was gathered from individuals within retail organisations in South Africa to explore the indirect impact of AI capabilities on organisational performance. A total of 145 participants from South Africa's retail sector were surveyed, and their responses were analysed using structural equation modelling with AMOS/ SPSS. The results indicate that AI capabilities positively influence process automation, cognitive insight generation, cognitive engagement, and innovativeness. While both process automation and innovativeness positively correlate with organisational performance, it was observed that cognitive insights and cognitive engagement do not significantly affect organisational performance. These findings explain the essential resources comprising an AI capability and highlight the consequences of nurturing such capabilities on critical organisational activities, thereby influencing organisational performance
  • Thumbnail Image
    Item
    Addressing The 4IR Skills Gap for South Africa’s Economy
    (University of the Witwatersrand, Johannesburg, 2023) Bodibe, Lerato; Venter, Robert
    The widespread enthusiasm and hysteria for Artificial Intelligence (AI) todays not only encourages but forces us to approach the future with a combination of childlike aw and mature concerns. A new and evolving set of skills is vita and needed, because automation and robotisation powered by AI is simultaneously creating and eroding jobs. The so-called digitally enabled jobs, AI-driven jobs, indisputably needs a skilled cadre of graduates. This is especially true for the ICT sector in South Africa, which is plagued by a serious skills shortage. The research conducted for this MBA social entrepreneurship project led to the aim of establishing an eSkills institute, specifically targeting the youth in Orange Farm. This community was chosen for its demographic and social condition where it faces high levels of youth unemployment, drug abuse, and lack of access to educational facilities by the youth. The proposed business model offers a promising approach for the eSkills Institute to achieve both its social and financial goals, bringing meaningful change to disadvantaged communities through digital skills training. This would enable the eSkills institute to achieve long-term viability and make a positive impact with its underlying primary objective of providing digital skills training to disadvantaged communities, thereby bridging the digital divide and increasing their access to economic opportunities. Basing our analysis on market research and stakeholder engagement has helped identify key areas of opportunity to generate revenue and create social value. These opportunities include offering paid digital skills training programs to corporate clients, partnering with government agencies to provide subsidised training to low-income individuals, and establishing a social enterprise arm that offers software development and design services to small businesses and their ecosystems
  • Thumbnail Image
    Item
    Perceptions and Adoption Trends of Artificial Intelligence in Portfolio Construction and Management in the Financial Services Industry of South Africa
    (University of the Witwatersrand, Johannesburg, 2024) Agjee, Zeyn; Horney, Sylvester
    The adoption of Artificial intelligence (AI) in portfolio construction promises to revolutionise financial services, offering opportunities to enhance efficiency, foster innovation, and drive disruptive change. This qualitative study investigates the perspectives and adoption trends of AI-driven portfolio construction methods among South African financial services organisations. The research uncovers attitudes, challenges, and aspirations surrounding AI adoption through in-depth interviews with nine industry professionals. The study finds that while there is widespread enthusiasm within the industry for AI adoption in portfolio construction and the industry, professionals express reservations about trust, lack of understanding, data challenges, costs and AI's efficacy in navigating the complexities of the South African market. The study highlights complexities in AI adoption, including transparency, regulatory compliance, accountability, data considerations, overfitting issues, human-machine interactions, lack of agility in companies and potential job displacement concerns. Despite the increasing acceptance of AI in investment management, significant obstacles persist, necessitating concerted efforts to address problems and cultivate trust and openness within the industry. The paper presents valuable insights into the patterns of AI adoption in South Africa, offering practical recommendations for industry practitioners, policymakers, and researchers. It emphasises the importance of trust-building strategies among industry practitioners, highlighting the need for transparent communication and ethical considerations throughout AI adoption. Additionally, the paper underscores the role of policymakers in developing regulatory frameworks that promote responsible AI integration, advocating for guidelines that uphold ethical principles and protect consumer rights Further, there is a call for continued support for research and development efforts tailored to the South African market, aiming to address specific challenges and foster innovation in AI technologies. Overall, the findings emphasise the necessity of collaboration between stakeholders to ensure the ethical and widespread practical adoption of AI in South Africa's financial sector
  • Thumbnail Image
    Item
    Rationalization of Deep Neural Networks in Credit Scoring
    (University of the Witwatersrand, Johannesburg, 2023-07) Dastile, Xolani Collen; Celik, Turgay
    Machine learning and deep learning, which are subfields of artificial intelligence, are undoubtedly pervasive and ubiquitous technologies of the 21st century. This is attributed to the enhanced processing power of computers, the exponential growth of datasets, and the ability to store the increasing datasets. Many companies are now starting to view their data as an asset, whereas previously, they viewed it as a by-product of business processes. In particular, banks have started to harness the power of deep learning techniques in their day-to-day operations; for example, chatbots that handle questions and answers about different products can be found on banks’ websites. One area that is key in the banking sector is the credit risk department. Credit risk is the risk of lending money to applicants and is measured using credit scoring techniques that profile applicants according to their risk. Deep learning techniques have the potential to identify and separate applicants based on their lending risk profiles. Nevertheless, a limitation arises when employing deep learning techniques in credit risk, stemming from the fact that these techniques lack the ability to provide explanations for their decisions or predictions. Hence, deep learning techniques are coined as non-transparent models. This thesis focuses on tackling the lack of transparency inherent in deep learning and machine learning techniques to render them suitable for adoption within the banking sector. Different statistical, classic machine learning, and deep learning models’ performances were compared qualitatively and quantitatively. The results showed that deep learning techniques outperform traditional machine learning models and statistical models. The predictions from deep learning techniques were explained using state-of-the-art explanation techniques. A novel model-agnostic explanation technique was also devised, and credit-scoring experts assessed its validity. This thesis has shown that different explanation techniques can be relied upon to explain predictions from deep learning and machine learning techniques.
  • Thumbnail Image
    Item
    Future Proofing Architecture: Intelligent design processes of an AI-Innovation center in Newtown
    (University of the Witwatersrand, Johannesburg, 2024-02) Wilson, Liam Robin; Triana-Martinez, Gustavo
    In the ever-evolving realm of architecture, tools used by architects and related professionals have transitioned from rudimentary sketches to sophisticated digital simulations. Today, the integration of Artificial Intelligence (AI) stands to redefine this lineage of tools, offering both challenges and opportunities. This thesis delves into AI’s transformative potential in architectural design processes specifically within a South African context, exploring its influence from conceptualization to the final design stages. Through a systematic methodology, the research herein investigates and compares the conventional architectural design stages, the current state of AI and its practical applications in architecture. I have carefully selected a handful of AI-driven software tools that have been instrumental in forging a generative design process. Central to this exploration, is the design of an AI Innovation Centre for Witwatersrand University in Newtown, Johannesburg. This Centre is not just a building but a manifestation of my core argument: that AI, when understood as a tool in the architect’s evolving toolkit, can profoundly influence design outcomes in a manner that far outreaches human capabilities. This study further importantly addresses the ethical implications of AI in architecture, advocating for a collaborative approach that not only complements human expertise, but that illustrates the pitfalls and certain biases inherent to AI. Through this comprehensive exploration, this thesis underscores the need for architectural spaces to evolve in response to AI-driven operational changes, while ensuring designs remain rooted in human-centric principles.