Influence of gap length on GAN-imputed missing dengue time series data

Authors

  • Andrea Rose V. Franco ⋅ PH National Institute of Physics, University of the Philippines Diliman
  • Maricor N. Soriano ⋅ PH National Institute of Physics, University of the Philippines Diliman https://orcid.org/0000-0002-7373-8386

Abstract

Missing data in dengue surveillance systems pose a challenge for disease monitoring and response. In this paper, we explored the use of Generative Adversarial Networks (GANs) in imputing missing dengue time series data. GAN simulations were conducted using the real-world missing data distribution from OpenDengue dataset. We incorporated a bidirectional LSTM in the GAN architecture to capture temporal dependencies. An RMSE below 0.10 was achieved for short gap lengths. However, increased gap lengths were associated with higher RMSE values. Visual analysis further supported the deteriorated imputation at longer gap lengths, where imputed data struggled to follow the true data. Iteration loss suggests that this framework could still be enhanced for a more accurate missing dengue time-series data imputation.

Downloads

Issue

Entangled!
25-28 June 2025, National Institute of Physics, University of the Philippines Diliman

Please visit the SPP2025 activity webpage for more information on this year's Physics Congress.

Article ID

SPP-2025-1A-04

Section

Biological and Medical Physics

Published

2025-06-16

How to Cite

[1]
ARV Franco and MN Soriano, Influence of gap length on GAN-imputed missing dengue time series data, Proceedings of the Samahang Pisika ng Pilipinas 43, SPP-2025-1A-04 (2025). URL: https://proceedings.spp-online.org/article/view/SPP-2025-1A-04.