Analysis code: differentiating notified dengue from malaria in Kolkata from routine surveillance fields (leakage-audited machine learning, 2014-2025)
Analysis code for the study 'Differentiating notified dengue from malaria in Kolkata from routine surveillance fields: a leakage-audited machine-learning study, 2014-2025'. Includes data preparation, the leakage-safe benchmark (seven classifiers, five class-imbalance strategies, five random splits and a temporal split), the leakage audit, field ablations, calibration and decision-curve analyses, figures, ward-level spatial statistics, a synthetic example dataset, and derived ward-level result tables without patient information. The Kolkata Municipal Corporation line-lists and the licensed ward boundaries are not included.
Resource information
- Format
- zip
- Original repository
- 23516233
- DOI
- 10.5281/zenodo.23516233
- Credit basis
- Creators in Zenodo record metadata
- Access
- Open files available from the original Zenodo deposit
- Last checked
- Nov 5, 2026
- License
- mit-license
- Last record update
- Nov 5, 2026
About this resource
Analysis code for the study 'Differentiating notified dengue from malaria in Kolkata from routine surveillance fields: a leakage-audited machine-learning study, 2014-2025'. Includes data preparation, the leakage-safe benchmark (seven classifiers, five class-imbalance strategies, five random splits and a temporal split), the leakage audit, field ablations, calibration and decision-curve analyses, figures, ward-level spatial statistics, a synthetic example dataset, and derived ward-level result tables without patient information. The Kolkata Municipal Corporation line-lists and the licensed ward boundaries are not included.