Research at SPARK Institute
At SPARK Institute of Data Science and Applied Statistics, we are committed to advancing knowledge through innovative research in statistics and data science. Our faculty and students engage in cutting-edge projects that address real-world challenges and push the boundaries of what's possible with data.
Ongoing Research Projects
Climate Change Impact Modeling
ActiveDeveloping advanced statistical models to predict and analyze the impacts of climate change on public health, agriculture, and economic systems in East Africa.
Infectious Disease Forecasting
ActiveCreating machine learning models for early detection and prediction of infectious disease outbreaks using real-time data from multiple sources including social media and health records.
Big Data Analytics for Healthcare
PlanningLeveraging big data technologies to analyze healthcare datasets for improving patient outcomes, optimizing resource allocation, and enhancing decision-making in clinical settings.
Agricultural Data Science
ActiveApplying data science techniques to improve agricultural productivity through crop yield prediction, pest detection, and optimal resource allocation for smallholder farmers.
Quantifying Uncertainty in Dense Neural Networks Using Monte Carlo Dropout
ActiveDeveloping novel techniques to quantify uncertainty in deep neural network predictions using Monte Carlo dropout methods. This research aims to improve the reliability and interpretability of neural network models in critical applications.
A Survey on Machine Learning in Early Stroke Prediction
ReviewComprehensive survey and analysis of current machine learning approaches for early stroke prediction. This project evaluates various algorithms, datasets, and performance metrics to identify best practices and future research directions.
A Deep CNN-ViT Ensemble for Stroke Prediction
ActiveDeveloping an innovative ensemble model combining Convolutional Neural Networks (CNN) and Vision Transformers (ViT) for improved stroke prediction from medical imaging data. This hybrid approach aims to leverage the strengths of both architectures.
Faculty Publications
Machine Learning Approaches for COVID-19 Prediction in Uganda
Statistical Methods for Climate Change Impact Assessment in East Africa
Public Health Data Analytics: A Comprehensive Review
Big Data Applications in Healthcare: Challenges and Opportunities
Data-Driven Agriculture: Transforming Farming in Uganda
Quantifying Uncertainty in Dense Neural Networks Using Monte Carlo Dropout: Methods and Applications
A Survey on Machine Learning in Early Stroke Prediction: Current Trends and Future Directions
A Deep CNN-ViT Ensemble for Stroke Prediction: Improving Accuracy and Interpretability
Research Labs
Epidemiology & Public Health Lab
Focused on statistical modeling of infectious diseases and public health data analysis to improve healthcare outcomes.
- Disease surveillance systems
- Epidemic forecasting models
- Public health informatics
Climate & Environmental Data Lab
Dedicated to analyzing climate data and developing models for environmental impact assessment and prediction.
- Climate modeling systems
- Environmental data analysis
- Impact assessment tools
Big Data & Machine Learning Lab
Exploring large-scale data processing and developing cutting-edge machine learning algorithms for various applications.
- Distributed computing systems
- Deep learning frameworks
- Data visualization tools
Agricultural Data Science Lab
Applying data science techniques to solve agricultural challenges and improve farming practices through data-driven insights.
- Crop yield prediction
- Pest and disease detection
- Resource optimization models
Neural Networks & Uncertainty Lab
Specializing in advanced neural network architectures and uncertainty quantification methods for improved model reliability.
- Monte Carlo dropout methods
- Bayesian neural networks
- Uncertainty visualization tools