University of Technology, Mauritius

Data envelopment analysis (DEA) is a method for identifying the best practices among peer decision-making units (DMUs). Early DEA models were suited for a production possibility set of positive inputs and outputs. In real setups, some inputs and outputs may be negative. Several past studies have provided methods of dealing with negative data in DEA. Furthermore, the premise of early DEA models was that DMUs are homogeneous, though DMUs may exhibit a network structure with negative input and output values in practice. In this study, we presented semi-oriented radial measure (SORM) network DEA models to assess the system as a whole as well as the first and second Stages. The proposed method was compared with a network bi-directional SORM (DSORM) model. The targets obtained from the SORM network model revealed an improvement from the observed values, whereas some of the Stage 2 targets from the DSORM model revealed reductions from the observed values. An integrated SORM model was used to analyse the overall system at once; a higher discrimination of the DMUs was observed in the model. The models were applied to data from 13 insurance companies in Mauritius. The premium collection Stage (Stage 1) was observed to be the dominant Stage of the insurance companies.

Credit scoring is an essential tool used by financial institutions to estimate the creditworthiness of individuals and businesses. It represents a numerical evaluation of a borrower's likelihood to repay a loan and categorizes applicants into 'Good' and 'Bad' classes. Institutions rely on these scores for lending decisions, portfolio risk management, and loan conditions such as rates, limits, and repayment schedules. From a customer's perspective, a higher score provides access to better financial products, while a lower score restricts opportunities and increases costs. Twelve machine learning models were tuned and evaluated for credit scoring in this study. Moreover, detailed steps are provided for data preprocessing to allow for reproducibility and SHAP metrics were used to provide transparency to stakeholders. Models were implemented in Python, with CatBoost performing best with an F1-score of 0.88, followed by LightGBM with 0.87.

Achieving high-fidelity 3D reconstructions of clothed humans from a single image is pivotal for applications in virtual reality, gaming, and the fashion industry. However, the challenge of accurately reconstructing subjects in loose clothing and complex poses has yet to be fully addressed. To bridge this gap, we propose a novel hybrid approach that combines decoupled side-view features and rebalanced parametric body model prior to guide detailed 3D human reconstruction. This approach can handle loose clothing and unusual poses simultaneously. Specifically, guided by the SMPL-X-based parametric human body prior, we leverage a Transformer-based framework to effectively decouple side-view features from input images. This process significantly enhances the accuracy of implicit-function-based reconstruction for complex poses, enabling a more precise representation of human body postures. Furthermore, to better adapt to diverse clothing types and avoid overfitting to the training data mainly consisting of tight clothing, we introduce a rebalancing coefficient within a positional embedding-based strategy. This coefficient adjusts the model’s reliance on the parametric body prior, enhancing the ability to capture details of loose clothing. Consequently, the model can generate more reliable Signed Distance Function (SDF) values, which are essential for creating high-fidelity 3D clothed human bodies. Extensive experiments demonstrate superior performance in detailed representations for loose clothing and maintaining robust reconstruction of complex poses.

Diffusion-based generative models achieve high-fidelity image synthesis; however, controlling internal representations for abstract visual concepts remains challenging due to the ambiguity of textual descriptions. In this work, we propose a prompt-guided concept-vector learning framework for the controllable manipulation of such concepts without requiring external human-annotated image pairs, segmentation masks, identity labels, or manually annotated editing targets. The method introduces a learnable concept vector optimized in the bottleneck (mid-block) feature space of a pretrained Stable Diffusion U-Net, while keeping all pretrained model parameters frozen. A multi-prompt data generation strategy based on paired positive and neutral prompts provides weak semantic guidance for capturing the target concept direction and reducing dependence on a single prompt formulation. The learned vector is further applied in an image-to-image setting through controlled noise injection and concept-guided denoising, enabling the semantic modification of real images while preserving structural content. The concept strength is controlled by a scaling parameter γ, while the image-to-image noise strength is controlled by β, allowing for a practical balance between semantic modification and structural fidelity. Experiments are conducted on two main abstract concepts, perfect skin and peaceful lake, with additional qualitative analysis on subjective portrait-level concepts. Quantitative evaluation using SSIM, LPIPS, and CLIP similarity demonstrates that the proposed method improves semantic alignment while maintaining structural preservation compared with Stable Diffusion image-to-image baselines. A human preference study further shows that concept-injected outputs are preferred in 76.0% of responses for perfect skin and 85.7% for peaceful lake. Ablation studies further demonstrate the controllability and robustness of the proposed framework. Overall, the method provides a simple and parameter-efficient approach for interpretable concept-level manipulation in diffusion models.

Medical anatomy forms the prime foundation in the field of clinical medicine. Observational, descriptive, qualitative analysis using questionnaire was carried out for first MBBS students at the department of Anatomy. Preference for both traditional cadaveric teaching and modern teaching techniques were appreciated among the medical students. A statistically significant proportion of students (p < 0.001) believed that, small group interactive sessions and early clinical exposure were essential for coping with hybrid PBL curriculum. Data the necessity to adopt the different teaching modalities for implementation in the modern era with the aid of available technologies.