ike10/Malaria_Visualization

Visualizing Over a Century of Malaria Mosquito Data Across Africa 🦟🌍 Explore trends in Anopheles mosquito species from 1898–2016 with data science and machine learning. Includes interactive maps, predictive models, and actionable insights to support global malaria control strategies. # 🦟 Malaria Mosquito Database (1898–2016) This project analyzes over a century’s worth of mosquito vector data across Africa, with a focus on *Anopheles* species—key vectors for malaria. It aims to identify geographic, temporal, and environmental patterns, and apply machine learning to aid global health strategies for malaria control. --- ## 🚀 Objectives - Understand the distribution of *Anopheles* mosquito species across Africa. - Reveal long-term sampling trends (1898–2016). - Visualize geospatial spread using interactive maps. - Predict mosquito presence using machine learning. - Identify high-impact zones for health intervention and funding prioritization. --- ## 🧪 Tools & Libraries - **Data Analysis**: `pandas`, `numpy` - **Visualization**: `matplotlib`, `seaborn`, `folium` - **Machine Learning**: `scikit-learn` (Random Forest Classifier, Permutation Importance) - **Geospatial Mapping**: `folium`, `MarkerCluster`, `HeatMap` --- ## 📊 Project Overview ### 1. Data Exploration - Displays dataset structure: column names, types, and missing values. - Summarizes statistical distribution of values. - Identifies key columns with missing data. ### 2. Top Species Visualization - Horizontal bar chart of the most frequently observed *Anopheles* species. - Highlights **An. gambiae** as the most common malaria vector. ### 3. Temporal Sampling Trends - Line graph illustrating sampling frequency from **1898 to 2016**. - Peak sampling activity occurs in the 2000s. ### 4. Geospatial Visualization - Interactive **Folium map** showing sample locations across Africa. - Color-coded markers representing vector species and countries. --- ## 🤖 Machine Learning Models ### 🔬 Example 1: Predicting *An. gambiae* Presence - **Model**: Random Forest Classifier - **Top Predictors**: - Latitude - Longitude - Year of sampling - **Visualizations**: - ROC curve - Confusion matrix - Feature importance plots ### 🧭 Example 2: WHO Intervention Decision Model - Predicts **WHO mal …