A Machine Learning System that aims at predicted the future malaria cases based on climatic conditions recorded in the different counties in Kenya.
# π¦ Malaria Case Prediction in Kenya Counties
## π― Problem Definition: Why This Project Exists
Malaria transmission in Kenya is highly seasonal, geographically uneven, and climate-driven. Counties experience spikes at different times of the year depending on rainfall, temperature, and historical transmission patterns.
## π Project Overview
**Malaria Case Prediction in Kenya Counties** is a data-driven machine learning project that predicts monthly malaria case counts for each county in Kenya using historical case data and climatic features such as temperature and rainfall.
This project demonstrates:
- Data acquisition, cleaning, and preprocessing
- Feature engineering for time series and climate data
- Model training and evaluation
π The goal of the project is to aid in prior planning by the stakeholders (MoH: Malaria Response Team). Through case prediction, the stakeholders are aware of areas that could have potential spikes in cases in a give time of the year and hence help in resource allocation to aid prevention.
---
## π Motivation
Malaria remains a major public health challenge in Kenya. Accurate short-term predictions can help:
- Health authorities allocate resources more effectively
- Counties prepare for expected case surges
- Non-governmental partners plan targeted interventions
This project focuses on predictive modeling using real case counts and weather data to provide actionable forecasting insights.
---
## π§ What This Project Does (At a High Level)
This project builds a county-level, monthly malaria case prediction system using machine learning.
Specifically, it:
1. Learns patterns from past malaria case counts
2. Incorporates climatic drivers (rainfall and temperature)
3. Produces forward-looking predictions for each county
It is a decision-support system.
### ποΈ Project Structure
```
C:.
β .gitignore
β requirements.txt
β
ββββconfig
β config.yaml
β
ββββdata
β malaria_raw.csv
β nyandarua.csv
β process_mala β¦