Hybrid machine learning approach to zero-inflated data improves accuracy of dengue prediction

Background
Spatiotemporal dengue forecasting using machine learning (ML) can contribute to the development of prevention and control strategies for impending dengue outbreaks. However, training data for dengue incidence may be inflated with frequent zero values because of the rarity of cases, which lowers the prediction accuracy. This study aimed to understand the influence of spatiotemporal resolutions of training data on the accuracy of dengue incidence prediction using ML models, to understand how the influence of spatiotemporal resolution differs between quantitative and qualitative predictions of dengue incidence, and to improve the accuracy of dengue incidence prediction with zero-inflated data.
Methodology
We predicted dengue incidence at six spatiotemporal resolutions and compared their prediction accuracy. Six ML algorithms were compared: generalized additive models, random forests, conditional inference forest (CIF), artificial neural networks, support vector machines and regr..., , , # Weekly Dengue Incidence and Environmental Data 2009 - 2013

[https://doi.org/10.5061/dryad.x3ffbg7ss](https://doi.org/10.5061/dryad.x3ffbg7ss)

This dataset contains weekly log transformed weekly dengue incidence and environmental data for each village in Metropolitan Manila, Philippines from January 2009 to December 2013.

## Description of the data and file structure

### Weekly_dengue_incidence_env_data.csv

Data columns are abbreviated and below are the descriptions

| **Colum name** | **Variable name** |
| :------------- | :---------------------------------------------------- |
| ID | Numeric identification of the observation |
| Year | Year of the observed data |
| week | Week of the observed data |
| City | City name of the observation |
| Village | Village name of the observation ...,