electricsheepafrica/africa-kenya-natural-disaster-inventory-mapped-by-event-type-1999-2013

---
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
- affected-population
- natural-disasters
- ken
pretty_name: "Kenya - Natural Disaster Inventory Mapped by Event Type"
dataset_info:
splits:
- name: train
num_examples: 1096
- name: test
num_examples: 274
---

# Kenya - Natural Disaster Inventory Mapped by Event Type

**Publisher:** Kenya Open Data Initiative (inactive) · **Source:** [HDX](https://data.humdata.org/dataset/kenya-natural-disaster-inventory-mapped-by-event-type-1999-2013) · **License:** `other-pd-nr` · **Updated:** 2023-03-03

---

## Abstract

The National Disaster inventory is a record of Natural Disasters including floods, thunderstorms, forest fires, mudslides and disease outbreaks etc. The inventory keeps track of the losses of life destruction of property and infrastructure, injury and displacement due to these incidents.

Each row in this dataset represents subnational administrative unit observations. Data was last updated on HDX on 2023-03-03. Geographic scope: **KEN**.

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

---

## Dataset Characteristics

| | |
|---|---|
| **Domain** | Epidemiology and disease surveillance |
| **Unit of observation** | Subnational administrative unit observations |
| **Rows (total)** | 1,371 |
| **Columns** | 18 (7 numeric, 11 categorical, 0 datetime) |
| **Train split** | 1,096 rows |
| **Test split** | 274 rows |
| **Geographic scope** | KEN |
| **Publisher** | Kenya Open Data Initiative (inactive) |
| **HDX last updated** | 2023-03-03 |

---

## Variables

**Geographic** — `county` (TANA RIVER, KITUI, MAKUENI), `district` (TANA RIVER, MAKUENI, NAIROBI), `place` (Timboiywo area, Turkana East, Sololo), `x` (range -4.4081–4.4756), `y` (range 34.0033–41.8148) and 2 others.

**Outcome / Measurement** — `affected` (range 1.0–334000.0).

**Identifier / Metadata** — `objectid` (range 1.0–1371.0), `esa_source` (HDX), `esa_processed`.

**Other** — `event` (DROUGHT, FLOOD, EPIDEMIC), `division` (GARSEN, YATTA, CENTRAL NAIROBI), `losses_usd` (range 0.0–0.0), `losses_local` (range 0.0–0.0), `hospitals` (yes, Yes) and 2 others.

---

## Quick Start

```python
from datasets import load_dataset

ds = load_dataset("electricsheepafrica/africa-kenya-natural-disaster-inventory-mapped-by-event-type-1999-2013")
train = ds["train"].to_pandas()
test = ds["test"].to_pandas()

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

---

## Schema

| Column | Type | Null % | Range / Sample Values |
|---|---|---|---|
| `objectid` | int64 | 0.0% | 1.0 – 1371.0 (mean 686.0) |
| `event` | object | 0.0% | DROUGHT, FLOOD, EPIDEMIC |
| `county` | object | 0.0% | TANA RIVER, KITUI, MAKUENI |
| `district` | object | 3.6% | TANA RIVER, MAKUENI, NAIROBI |
| `division` | object | 24.5% | GARSEN, YATTA, CENTRAL NAIROBI |
| `place` | object | 26.1% | Timboiywo area, Turkana East, Sololo |
| `affected` | float64 | 68.6% | 1.0 – 334000.0 (mean 17344.0907) |
| `losses_usd` | int64 | 0.0% | 0.0 – 0.0 (mean 0.0) |
| `losses_local` | float64 | 0.7% | 0.0 – 0.0 (mean 0.0) |
| `hospitals` | object | 70.4% | yes, Yes |
| `lost_cattle` | float64 | 1.1% | 0.0 – 0.0 (mean 0.0) |
| `comments` | object | 76.2% | Incidences of cholera were reported as from January to June 2009, Indigenous forest burned down, People affected are located in the relocation camps |
| `x` | float64 | 0.1% | -4.4081 – 4.4756 (mean -0.3781) |
| `y` | float64 | 0.1% | 34.0033 – 41.8148 (mean 37.2112) |
| `constituency_gis` | object | 0.0% | GALOLE, KITUI EAST, KIBWEZI WEST |
| `county_centriod` | object | 0.0% | (-1.52636660432, 39.41786920290), (-1.49103853075, 38.40692101580), (-2.15768410007, 37.78824557250) |
| `esa_source` | object | 0.0% | HDX |
| `esa_processed` | object | 0.0% | |

---

## Numeric Summary

| Column | Min | Max | Mean | Median |
|---|---|---|---|---|
| `objectid` | 1.0 | 1371.0 | 686.0 | 686.0 |
| `affected` | 1.0 | 334000.0 | 17344.0907 | 1184.0 |
| `losses_usd` | 0.0 | 0.0 | 0.0 | 0.0 |
| `losses_local` | 0.0 | 0.0 | 0.0 | 0.0 |
| `lost_cattle` | 0.0 | 0.0 | 0.0 | 0.0 |
| `x` | -4.4081 | 4.4756 | -0.3781 | -0.5107 |
| `y` | 34.0033 | 41.8148 | 37.2112 | 37.0234 |

---

## 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`. 27 column(s) with >80% missing values were removed: `date`, `deaths`, `injured`, `missing`, `houses_destroyed`, `houses_damaged`.... 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 Kenya Open Data Initiative (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: `division`, `place`, `affected`, `hospitals`, `comments`.
- Refer to the [original HDX dataset page](https://data.humdata.org/dataset/kenya-natural-disaster-inventory-mapped-by-event-type-1999-2013) for the publisher's own methodology notes and caveats.

---

## Citation

```bibtex
@dataset{hdx_africa_kenya_natural_disaster_inventory_mapped_by_event_type_1999_2013,
title = {Kenya - Natural Disaster Inventory Mapped by Event Type},
author = {Kenya Open Data Initiative (inactive)},
year = {2023},
url = {https://data.humdata.org/dataset/kenya-natural-disaster-inventory-mapped-by-event-type-1999-2013},
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.*