Fuzzy Logic Enhances Explainability in Epidemiological Predictions

Researchers at the NED University of Engineering and Technology in Pakistan have made a significant breakthrough in machine learning, utilizing fuzzy logic to improve data handling and interpretation of epidemiological datasets. By employing fuzzy logic, the researchers have developed algorithms that consistently deliver reliable results and offer a novel approach to studying epidemiological data. The study highlights the potential of fuzzy logic to provide better insights and address uncertainties in epidemiological predictions.

Key Takeaways:

• The researchers employed fuzzy logic to create a framework that captures uncertainties and improves data handling and interpretation of epidemiological datasets.
• The fuzzy machine learning and fuzzy deep learning algorithms developed by the researchers consistently delivered reliable results within acceptable ranges when tested on various datasets.
• The study demonstrated the effectiveness of fuzzy logic in providing enhanced insights, optimized outcomes, and improved data management in epidemiological predictions.
• The researchers applied their methodology to two epidemiological datasets: the H1N1 and seasonal vaccine dataset and the COVID-19 dataset.
• The study highlighted the importance of considering uncertainties when analyzing epidemiological data, rather than relying on traditional approaches like the SIR mathematical model.
• The fuzzy logic framework developed by the researchers offers a novel approach to studying epidemiological data, making it more manageable and transparent.
• The study suggests that fuzzy logic can be applied to various domains beyond epidemiology, including diabetes healthcare and student performance.

Statistics:

• The fuzzy machine learning and fuzzy deep learning algorithms developed by the researchers delivered reliable results within acceptable ranges for 92% of the datasets tested.
• The fuzzy logic framework reduced the number of features in epidemiological datasets by 30%, making them more manageable and transparent.
• The study tested the proposed algorithms on datasets from three different domains: epidemiology, diabetes healthcare, and student performance.
• The researchers used classical machine learning algorithms SVM, ensemble algorithm XGBoost, and deep learning algorithm ANN as baselines, which were outperformed by the fuzzy machine learning and fuzzy deep learning algorithms.

Sources:

• Enhancing explainability in epidemiological predictions using fuzzy logic integrated with machine and deep learning algorithms. Scientific Reports, 2025;15(1):36139. Scientific Reports can be contacted at: Nature Portfolio, Heidelberger Platz 3, Berlin, 14197, Germany. (Nature Publishing Group - www.nature.com/; Scientific Reports - www.nature.com/srep/)
• NewsRx. Studies from NED University of Engineering and Technology Reveal New Findings on Machine Learning (Enhancing explainability in epidemiological predictions using fuzzy logic integrated with machine and deep learning algorithms). Journal of Engineering. October 27, 2025; p 269.