Conference Publications

Permanent URI for this collection

http://hdl.handle.net/10204/11486

Browse

Browsing Conference Publications by browse.metadata.impactarea "Artificial Intel & Extel Reality"

Now showing 1 - 2 of 2
  • No Thumbnail Available
    Item
    Investigating gender bias using artificial intelligence classification models on RAVDESS dataset
    (2025-08) Sefara,Tshephisho J; Khosa, Marshal V; Kisten, Melvin
    Artificial intelligence (AI) classification models are increasingly being deployed across a wide array of sectors, becoming fundamental tools in decision-making processes that impact individuals and society. These models are utilised in critical applications such as healthcare diagnostics, financial risk assessment, criminal justice systems, and educational admissions, demonstrating their widespread influence. However, a significant challenge arises from the susceptibility of these models to biases, which can lead to outcomes that are unfair, discriminatory, and ultimately harmful to individuals and specific demographic groups. Artificial intelligence bias refers to systematic errors that occur within decision-making processes, ultimately leading to outcomes that are unfair or inequitable. This can manifest as skewed results stemming from human biases that have influenced the original data used to train the AI model, resulting in distorted outputs with potentially harmful results. In this paper, we mitigate the gender bias that occurred during data selection for a classification model. This research experiment was conducted on RAVDESS emotion recognition dataset. The experiments showed improvement in model accuracy by 6% after bias mitigation.
  • No Thumbnail Available
    Item
    XR-training: A mixed reality platform for accelerated industrial equipment training
    (2025-07) Moodley, Jayandren; Van Eden, Beatrice; Mphephu, Mutali; Dire, Patrick OM
    Extended Reality (XR) technologies, including Augmented Reality (AR) and Mixed Reality (MR), are transforming industrial training by enabling immersive, scalable skill development. This paper presents a deployed framework integrating a Content Management System (CMS) for 3D/AR asset sustainability and Azure Communication Services (ACS) for remote collaboration, tailored to address South Africa’s digital resilience challenges. The platform leverages hierarchical asset categorization in the CMS to reduce redundancy, achieving a 90% reduction in duplicate 3D assets, while ACS-enabled remote guidance ensures 95% AR tracking accuracy during equipment training sessions. Evaluation metrics demonstrate a 60% decrease in on-site technical visits, lowering CO2 emissions, and a 40% improvement in trainee competency scores (measured via pre-/post-assessments). While leveraging commercial tools (Microsoft Teams, SharePoint), the framework’s novelty lies in its adaptation to infrastructure constraints (e.g., low-bandwidth optimization) and localized use cases (e.g., digital twin integration for legacy machinery). Deployment outcomes from a pilot with 150 trainees highlight scalability, with 85% reporting reduced dependency on physical trainers. This work aligns with 4IR goals by combining technological feasibility (validated ACS/CMS performance) with socio-economic impact, positioning XR as a catalyst for equitable digital economies.