electricsheepafrica/africa-cote-divoire-healthsites

---
annotations_creators:
- no-annotation
language_creators:
- found
language:
- en
license: cc-by-4.0
multilinguality:
- monolingual
size_categories:
- 1K<n<10K
source_datasets:
- original
task_categories:
- tabular-classification
task_ids: []
tags:
- africa
- humanitarian
- hdx
- electric-sheep-africa
- geodata
- health-facilities
- civ
pretty_name: "Côte d'Ivoire-healthsites"
dataset_info:
splits:
- name: train
num_examples: 1334
- name: test
num_examples: 333
---

# Côte d'Ivoire-healthsites

**Publisher:** Global Healthsites Mapping Project · **Source:** [HDX](https://data.humdata.org/dataset/cote-divoire-healthsites) · **License:** `cc-by-igo` · **Updated:** 2025-04-25

---

## Abstract

This dataset shows the list of operating health facilities. Attributes included: Name,Nature of Facility, Activities, Lat, Long

Each row in this dataset represents tabular records. Temporal coverage is indicated by the `changeset_timestamp` column(s). Geographic scope: **CIV**.

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

---

## Dataset Characteristics

| | |
|---|---|
| **Domain** | Public health |
| **Unit of observation** | Tabular records |
| **Rows (total)** | 1,668 |
| **Columns** | 15 (6 numeric, 8 categorical, 1 datetime) |
| **Train split** | 1,334 rows |
| **Test split** | 333 rows |
| **Geographic scope** | CIV |
| **Publisher** | Global Healthsites Mapping Project |
| **HDX last updated** | 2025-04-25 |

---

## Variables

**Geographic** — `x` (range -8.2954–-2.7927), `y` (range 4.6541–9.6108), `osm_type` (node, way), `amenity` (pharmacy, doctors, hospital), `operator_type` (private, public, ngo).

**Temporal** — `changeset_timestamp`.

**Identifier / Metadata** — `osm_id` (range -4034977.0–6781232575.0), `changeset_id` (range 4955551.0–75313842.0), `uuid` (ee57a432a8d94e5fabd7e7c7a29748c7, 1c9dae47930d4967ba489353569439de, c1e434e2fdd74f45aa8cbd089cd83c84), `name` (Clinique Chinoise, Infirmerie, Clinique), `esa_source` (HDX) and 1 others.

**Other** — `completeness` (range 6.0–31.0), `changeset_version` (range 1.0–10.0), `changeset_user` (sommerluk, anebophil, ulrichm).

---

## Quick Start

```python
from datasets import load_dataset

ds = load_dataset("electricsheepafrica/africa-cote-divoire-healthsites")
train = ds["train"].to_pandas()
test = ds["test"].to_pandas()

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

---

## Schema

| Column | Type | Null % | Range / Sample Values |
|---|---|---|---|
| `x` | float64 | 4.5% | -8.2954 – -2.7927 (mean -4.7813) |
| `y` | float64 | 4.5% | 4.6541 – 9.6108 (mean 5.8709) |
| `osm_id` | int64 | 0.0% | -4034977.0 – 6781232575.0 (mean 4949679221.6535) |
| `osm_type` | object | 0.0% | node, way |
| `completeness` | int64 | 0.0% | 6.0 – 31.0 (mean 12.5929) |
| `amenity` | object | 0.6% | pharmacy, doctors, hospital |
| `changeset_id` | float64 | 0.1% | 4955551.0 – 75313842.0 (mean 58441148.0978) |
| `uuid` | object | 0.0% | ee57a432a8d94e5fabd7e7c7a29748c7, 1c9dae47930d4967ba489353569439de, c1e434e2fdd74f45aa8cbd089cd83c84 |
| `changeset_version` | float64 | 0.1% | 1.0 – 10.0 (mean 1.904) |
| `changeset_timestamp` | datetime64[ns] | 0.1% | |
| `name` | object | 4.5% | Clinique Chinoise, Infirmerie, Clinique |
| `changeset_user` | object | 0.1% | sommerluk, anebophil, ulrichm |
| `operator_type` | object | 76.5% | private, public, ngo |
| `esa_source` | object | 0.0% | HDX |
| `esa_processed` | object | 0.0% | 2026-04-07 |

---

## Numeric Summary

| Column | Min | Max | Mean | Median |
|---|---|---|---|---|
| `x` | -8.2954 | -2.7927 | -4.7813 | -4.0811 |
| `y` | 4.6541 | 9.6108 | 5.8709 | 5.3684 |
| `osm_id` | -4034977.0 | 6781232575.0 | 4949679221.6535 | 5568888182.5 |
| `completeness` | 6.0 | 31.0 | 12.5929 | 10.0 |
| `changeset_id` | 4955551.0 | 75313842.0 | 58441148.0978 | 62691687.0 |
| `changeset_version` | 1.0 | 10.0 | 1.904 | 2.0 |

---

## 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`. 21 column(s) with >80% missing values were removed: `is_in_health_zone`, `speciality`, `addr_full`, `operator`, `water_source`, `insurance`.... 1 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 Global Healthsites Mapping Project 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: `operator_type`.
- Refer to the [original HDX dataset page](https://data.humdata.org/dataset/cote-divoire-healthsites) for the publisher's own methodology notes and caveats.

---

## Citation

```bibtex
@dataset{hdx_africa_cote_divoire_healthsites,
title = {Côte d'Ivoire-healthsites},
author = {Global Healthsites Mapping Project},
year = {2025},
url = {https://data.humdata.org/dataset/cote-divoire-healthsites},
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.*