Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{181922,
        author = {Rasheeja A P},
        title = {Epidemic Forecasting at a Crossroads: Machine Learning vs. Deep Learning- Accuracy, Speed, and Practical Trade-offs},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {12},
        number = {2},
        pages = {148-155},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=181922},
        abstract = {Predicting epidemic outbreaks accurately is crucial for effective public health planning, helping authorities act quickly and allocate resources wisely. In this study, we compare five different forecasting models—SARIMAX, XGBoost, LSTM-Pro, Transformer-TS, and N-BEATS—to see how well they perform over different timeframes: 7-day, 14-day, and 30-day forecasts. Using real-world epidemic data, we assess each model's accuracy with key metrics like Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), and Mean Absolute Percentage Error (MAPE), while also considering how efficiently they run. The results show that the Transformer-TS model delivers the most accurate predictions, with an MAE of 275 and RMSE of 385. However, it takes much longer to train—nearly 4.8 hours. On the other hand, SARIMAX is much faster, training in just 0.2 hours, though it sacrifices some accuracy.
This research highlights the trade-offs between accuracy, speed, and ease of interpretation, providing public health officials with practical guidance. Based on these findings, we offer tailored recommendations for choosing the right model depending on the outbreak situation and available resources.},
        keywords = {deep learning, epidemic forecasting, machine learning, performance evaluation, prediction, outbreak},
        month = {June},
        }