Pascal SUNGU

Personal academic and professional site of Pascal Sungu.

đŸ§Ș PhD Project – University of Douala (Ongoing)

Title: Evaluate the Effectiveness and Cost-Effectiveness of Long-lasting Indoor Residual Spraying for Malaria Control in the DR Congo: A Mathematical Modelling Approach

Abstract:
Malaria remains a major public health threat in the DRC, responsible for over 12% of global cases. Despite increased coverage of ITNs and IRS, the effectiveness of interventions is often limited by resistance and poor implementation. This project uses hierarchical Bayesian models to quantify IRS efficacy in different ecological settings, and integrates these findings into a calibrated malaria transmission model. The goal is to recommend scalable, cost-effective IRS deployment strategies specific to DRC.

🔐 MSc Project – PAUSTI (2021)

Title: A Mathematical Modeling Approach in Cybersecurity Using Deep Neural Learning

Abstract:
Inspired by epidemiological models, this project introduces a SEIARS model (Susceptible–Exposed–Infected–Asymptomatic–Recovered–Susceptible) to study malware dynamics in computer networks. A deep neural network approach was used to approximate the solution of the model’s differential equations, improving both memory efficiency and approximation quality compared to traditional finite difference methods. The BFGS optimization method yielded the best convergence performance.

🧠 AIMS Essay Project – AIMS Cameroon (2018)

Title: Applications of Concept Lattice Theory in Data Mining and Image Analysis

Abstract:
This essay applies Formal Concept Analysis (FCA) — a lattice-theoretic tool — to extract frequent co-occurring patterns in datasets. It demonstrates how concept lattices can be extended to handle image dataset characterization using landmarks. The work supports knowledge discovery in structured and semi-structured data environments.