electricsheepafrica/africa-darfur-damaged-and-destroyed-villages

---
annotations_creators:
- no-annotation
language_creators:
- found
language:
- en
license: other
multilinguality:
- monolingual
size_categories:
- 1K<n<10K
source_datasets:
- original
task_categories:
- tabular-classification
- other
task_ids: []
tags:
- africa
- humanitarian
- hdx
- electric-sheep-africa
- conflict-violence
- eastern-africa
- geodata
- sdn
pretty_name: "Darfur Damaged and Destroyed Villages"
dataset_info:
splits:
- name: train
num_examples: 7346
- name: test
num_examples: 1836
---

# Darfur Damaged and Destroyed Villages

**Publisher:** U.S. Department of State - Humanitarian Information Unit (inactive) · **Source:** [HDX](https://data.humdata.org/dataset/darfur-damaged-and-destroyed-villages) · **License:** `other-pd-nr` · **Updated:** 2025-05-05

---

## Abstract

The Darfur Damaged and Destroyed Villages dataset describes the condition of villages in the Darfur region of Sudan that the U.S. Government has confirmed as either damaged or destroyed between the time period February 2003 to December 2010. Additionally, villages that are confirmed to have No Damage are also reported.

Each row in this dataset represents geolocated point observations. Data was last updated on HDX on 2025-05-05. Geographic scope: **SDN**.

*Curated into ML-ready Parquet format by [Electric Sheep Africa](https://huggingface.co/electricsheepafrica).*

---

## Dataset Characteristics

| | |
|---|---|
| **Domain** | Conflict and security |
| **Unit of observation** | Geolocated point observations |
| **Rows (total)** | 9,183 |
| **Columns** | 14 (9 numeric, 5 categorical, 0 datetime) |
| **Train split** | 7,346 rows |
| **Test split** | 1,836 rows |
| **Geographic scope** | SDN |
| **Publisher** | U.S. Department of State - Humanitarian Information Unit (inactive) |
| **HDX last updated** | 2025-05-05 |

---

## Variables

**Geographic** — `latitude` (range 91624.0–14243299.0), `lat_dd` (range 9.2733–15.5889), `longitude` (range 215006.0–25534651.0), `long_dd` (range 21.835–27.39), `yr_confirm` (range 2003.0–2010.0) and 4 others.

**Identifier / Metadata** — `name` (HASHABA, KUMA, HILLET IBRAHIM), `esa_source` (HDX), `esa_processed` (2026-04-05).

**Other** — `status` (NO DAMAGE, DESTROYED, DAMAGED), `structures` ( , ~25 of ~25, ~50 of ~50).

---

## Quick Start

```python
from datasets import load_dataset

ds = load_dataset("electricsheepafrica/africa-darfur-damaged-and-destroyed-villages")
train = ds["train"].to_pandas()
test = ds["test"].to_pandas()

print(train.shape)
train.head()
```

---

## Schema

| Column | Type | Null % | Range / Sample Values |
|---|---|---|---|
| `name` | object | 68.8% | HASHABA, KUMA, HILLET IBRAHIM |
| `latitude` | float64 | 1.5% | 91624.0 – 14243299.0 (mean 150722.5973) |
| `lat_dd` | float64 | 0.0% | 9.2733 – 15.5889 (mean 12.8066) |
| `longitude` | float64 | 1.5% | 215006.0 – 25534651.0 (mean 289943.1203) |
| `long_dd` | float64 | 0.0% | 21.835 – 27.39 (mean 24.4526) |
| `status` | object | 0.0% | NO DAMAGE, DESTROYED, DAMAGED |
| `structures` | object | 58.7% | , ~25 of ~25, ~50 of ~50 |
| `yr_confirm` | float64 | 71.8% | 2003.0 – 2010.0 (mean 2004.7153) |
| `aprox_str1` | float64 | 78.1% | 5.0 – 1500.0 (mean 105.8607) |
| `aprox_str2` | float64 | 70.5% | 10.0 – 4000.0 (mean 117.8581) |
| `yr_range1` | float64 | 66.3% | 2003.0 – 2010.0 (mean 2004.6276) |
| `yr_range2` | float64 | 66.3% | 2003.0 – 2010.0 (mean 2004.7904) |
| `esa_source` | object | 0.0% | HDX |
| `esa_processed` | object | 0.0% | 2026-04-05 |

---

## Numeric Summary

| Column | Min | Max | Mean | Median |
|---|---|---|---|---|
| `latitude` | 91624.0 | 14243299.0 | 150722.5973 | 125656.0 |
| `lat_dd` | 9.2733 | 15.5889 | 12.8066 | 12.9725 |
| `longitude` | 215006.0 | 25534651.0 | 289943.1203 | 243237.0 |
| `long_dd` | 21.835 | 27.39 | 24.4526 | 24.5558 |
| `yr_confirm` | 2003.0 | 2010.0 | 2004.7153 | 2004.0 |
| `aprox_str1` | 5.0 | 1500.0 | 105.8607 | 50.0 |
| `aprox_str2` | 10.0 | 4000.0 | 117.8581 | 75.0 |
| `yr_range1` | 2003.0 | 2010.0 | 2004.6276 | 2004.0 |
| `yr_range2` | 2003.0 | 2010.0 | 2004.7904 | 2004.0 |

---

## Curation

Raw data was downloaded from HDX via the CKAN API and converted to Parquet. Column names were lowercased and standardised to snake_case. Common missing-value markers (`N/A`, `null`, `none`, `-`, `unknown`, `no data`, `#N/A`) were unified to `NaN`. 1 column(s) with >80% missing values were removed: `sec_town`. 3 column(s) were cast from string to numeric or datetime based on parse-success rate (>85% threshold). The dataset was split 80/20 into train and test partitions using a fixed random seed (42) and saved as Snappy-compressed Parquet.

---

## Limitations

- Data originates from U.S. Department of State - Humanitarian Information Unit (inactive) and has not been independently validated by ESA.
- Automated cleaning cannot correct for misreported values, definitional inconsistencies, or sampling bias in the original collection.
- The following columns have >20% missing values and should be treated with caution in modelling: `name`, `structures`, `yr_confirm`, `aprox_str1`, `aprox_str2`, `yr_range1`, `yr_range2`.
- Refer to the [original HDX dataset page](https://data.humdata.org/dataset/darfur-damaged-and-destroyed-villages) for the publisher's own methodology notes and caveats.

---

## Citation

```bibtex
@dataset{hdx_africa_darfur_damaged_and_destroyed_villages,
title = {Darfur Damaged and Destroyed Villages},
author = {U.S. Department of State - Humanitarian Information Unit (inactive)},
year = {2025},
url = {https://data.humdata.org/dataset/darfur-damaged-and-destroyed-villages},
note = {Repackaged for machine learning by Electric Sheep Africa (https://huggingface.co/electricsheepafrica)}
}
```

---

*[Electric Sheep Africa](https://huggingface.co/electricsheepafrica) — Africa's ML dataset infrastructure. Lagos, Nigeria.*