ML system flagging overlapping and contested land parcels and prioritising dispute resolution to address Nigeria's land administration crisis.
# Land Tenure & Property Dispute Resolution System
GIS + ML system that digitises land parcel boundaries, flags overlapping or contested parcels, and prioritises resolution cases, tackling Nigeria's land administration crisis with data-driven dispute detection and resolution scheduling.
---
## Problem Statement
Nigeria's land administration system is severely fragmented, enabling rampant land grabbing, double allocation, and boundary disputes. This system provides automated dispute detection and resolution prioritisation to support land registries and courts.
---
## Features
| Feature | Description |
|---------|-------------|
| 3,000 Parcel Synthetic Dataset | Boundary, ownership, documentation, and dispute features |
| Gradient Boosting Dispute Model | Probability score per parcel |
| Resolution Urgency Scoring | Boundary overlap × area × prior disputes |
| Interactive Dispute Risk Map | Folium heatmap with per-parcel details |
| State & Land-Use Analysis | Dispute rate breakdown by state and land-use type |
---
## Tech Stack
| Layer | Technology |
|-------|-----------|
| Machine Learning | scikit-learn (Gradient Boosting) |
| Geospatial | GeoPandas, Folium |
| Data | pandas, NumPy |
| Visualisation | Matplotlib, Seaborn, Plotly |
---
## Project Structure
```
land-tenure-dispute/
├── src/
│ ├── data_loader.py # Parcel data generation and dispute risk classification
│ ├── model.py # GBM dispute predictor, resolution prioritisation
│ └── visualize.py # Dispute map and overlap analysis charts
├── data/raw/ # Land parcel records, ownership history
├── models/ # Saved model artifacts
├── config.yaml # Risk thresholds, state boundaries
├── main.py # Pipeline entry point
└── requirements.txt
```
---
## Quick Start
```bash
git clone
github.com
cd land-tenure-dispute
pip install -r requirements.txt
python main.py
```
---
## Data Sou …