electricsheepafrica/africa-south-sudan-humanitarian-access-incidents

---
annotations_creators:
- no-annotation
language_creators:
- found
language:
- en
license: cc-by-4.0
multilinguality:
- monolingual
size_categories:
- n<1K
source_datasets:
- original
task_categories:
- other
task_ids: []
tags:
- africa
- humanitarian
- hdx
- electric-sheep-africa
- humanitarian-access
- ssd
pretty_name: "South Sudan: Humanitarian Access"
dataset_info:
splits:
- name: train
num_examples: 49
- name: test
num_examples: 12
---

# South Sudan: Humanitarian Access

**Publisher:** OCHA South Sudan · **Source:** [HDX](https://data.humdata.org/dataset/south-sudan-humanitarian-access-incidents) · **License:** `cc-by` · **Updated:** 2024-01-15

---

## Abstract

The dataset contains information on the number of humanitarian access incidents in South Sudan. Humanitarian access concerns humanitarian actors’ ability to reach people affected by the crisis, as well as affected people’s ability to access humanitarian assistance and services.

Each row in this dataset represents subnational administrative unit observations. Data was last updated on HDX on 2024-01-15. Geographic scope: **SSD**.

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

---

## Dataset Characteristics

| | |
|---|---|
| **Domain** | Humanitarian and development data |
| **Unit of observation** | Subnational administrative unit observations |
| **Rows (total)** | 62 |
| **Columns** | 7 (1 numeric, 6 categorical, 0 datetime) |
| **Train split** | 49 rows |
| **Test split** | 12 rows |
| **Geographic scope** | SSD |
| **Publisher** | OCHA South Sudan |
| **HDX last updated** | 2024-01-15 |

---

## Variables

**Geographic** — `admin1` (Upper Nile, Jonglei, Unity), `admin1_code` (SS07, SS03, SS06), `admin2` (Abyei region, Juba, Mayendit), `admin2_code` (SS0001, SS0101, SS0605).

**Identifier / Metadata** — `incident` (range 1.0–392.0), `esa_source` (HDX), `esa_processed` (2026-04-10).

---

## Quick Start

```python
from datasets import load_dataset

ds = load_dataset("electricsheepafrica/africa-south-sudan-humanitarian-access-incidents")
train = ds["train"].to_pandas()
test = ds["test"].to_pandas()

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

---

## Schema

| Column | Type | Null % | Range / Sample Values |
|---|---|---|---|
| `admin1` | object | 1.6% | Upper Nile, Jonglei, Unity |
| `admin1_code` | object | 3.2% | SS07, SS03, SS06 |
| `admin2` | object | 3.2% | Abyei region, Juba, Mayendit |
| `admin2_code` | object | 3.2% | SS0001, SS0101, SS0605 |
| `incident` | float64 | 1.6% | 1.0 – 392.0 (mean 12.8525) |
| `esa_source` | object | 0.0% | HDX |
| `esa_processed` | object | 0.0% | 2026-04-10 |

---

## Numeric Summary

| Column | Min | Max | Mean | Median |
|---|---|---|---|---|
| `incident` | 1.0 | 392.0 | 12.8525 | 4.0 |

---

## 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`. 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 OCHA South Sudan and has not been independently validated by ESA.
- Automated cleaning cannot correct for misreported values, definitional inconsistencies, or sampling bias in the original collection.
- Refer to the [original HDX dataset page](https://data.humdata.org/dataset/south-sudan-humanitarian-access-incidents) for the publisher's own methodology notes and caveats.

---

## Citation

```bibtex
@dataset{hdx_africa_south_sudan_humanitarian_access_incidents,
title = {South Sudan: Humanitarian Access},
author = {OCHA South Sudan},
year = {2024},
url = {https://data.humdata.org/dataset/south-sudan-humanitarian-access-incidents},
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.*