How dhwanikaneria/dengueprediction works
Project type and purpose
• This is a machine‑learning batch‑processing pipeline whose goal is to predict dengue outbreaks in San Juan and Iquitos.
• Key indicators: a single Jupyter notebook (DenguePrediction.ipynb), a Data folder full of CSVs, and a final report.
File structure
• Top‑level items:
• Data/ (contains raw and submission‑format CSVs)
• DenguePrediction.ipynb (the interactive analysis and modeling script)
• Project_Final_Report.pdf (documentation of findings)
• README.txt (project description and setup)
• No separate “src” or “backend” folders – all logic lives in the notebook.
• No deployment or service configuration files – this is a research pipeline, not a running service.
README for additional insights
• Purpose: predict next‐week dengue cases using climate/vegetation features.
• Workflow:
• Load CSVs with Pandas
• Shift features by 1–3 weeks to capture incubation lag
• Train models in the notebook
• Evaluate via Mean Absolute Error (MAE)
• Dependencies: Python, Anaconda, Jupyter; likely Pandas, NumPy, scikit‑learn.