electricsheepafrica/africa-ucdp-data-for-ethiopia

---
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
- other
task_ids: []
tags:
- africa
- humanitarian
- hdx
- electric-sheep-africa
- conflict-violence
- hxl
- eth
pretty_name: "Ethiopia - Data on Conflict Events"
dataset_info:
splits:
- name: train
num_examples: 4308
- name: test
num_examples: 1077
---

# Ethiopia - Data on Conflict Events

**Publisher:** HDX · **Source:** [HDX](https://data.humdata.org/dataset/ucdp-data-for-ethiopia) · **License:** `cc-by-igo` · **Updated:** 2026-04-03

---

## Abstract

This dataset is UCDP's most disaggregated dataset, covering individual events of organized violence (phenomena of lethal violence occurring at a given time and place). These events are sufficiently fine-grained to be geo-coded down to the level of individual villages, with temporal durations disaggregated to single, individual days.
Sundberg, Ralph, and Erik Melander, 2013, “Introducing the UCDP Georeferenced Event Dataset”, Journal of Peace Research, vol.50, no.4, 523-532
Högbladh Stina, 2019, “UCDP GED Codebook version 19.1”, Department of Peace and Conflict Research, Uppsala University

Each row in this dataset represents first-level administrative unit observations. Temporal coverage is indicated by the `date_start`, `date_end` column(s). Geographic scope: **ETH**.

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

---

## Dataset Characteristics

| | |
|---|---|
| **Domain** | Conflict and security |
| **Unit of observation** | First-level administrative unit observations |
| **Rows (total)** | 5,386 |
| **Columns** | 51 (27 numeric, 21 categorical, 2 datetime) |
| **Train split** | 4,308 rows |
| **Test split** | 1,077 rows |
| **Geographic scope** | ETH |
| **Publisher** | HDX |
| **HDX last updated** | 2026-04-03 |

---

## Variables

**Geographic** — `year` (range 1989.0–2024.0), `active_year`, `type_of_violence` (range 1.0–3.0), `dyad_dset_id` (range 91.0–17936.0), `dyad_new_id` (range 555.0–17936.0) and 9 others.

**Temporal** — `source_date` (2016-06-30, 2023-12-26, 2022-10-31), `date_prec` (range 1.0–5.0), `date_start`, `date_end`.

**Outcome / Measurement** — `number_of_sources` (range -1.0–22.0), `deaths_a` (range 0.0–121848.0), `deaths_b`, `deaths_civilians`, `deaths_unknown`.

**Identifier / Metadata** — `id` (range 7340.0–568501.0), `relid` (ETH-2020-1-555-0, ETH-2000-2-193-1.4, ETH-2018-2-5349-0), `code_status` (Clear), `conflict_dset_id` (range 91.0–17936.0), `conflict_new_id` (range 267.0–16235.0) and 14 others.

**Other** — `where_prec` (range 1.0–6.0), `where_description`, `adm_1`, `adm_2`, `geom_wkt` and 4 others.

---

## Quick Start

```python
from datasets import load_dataset

ds = load_dataset("electricsheepafrica/africa-ucdp-data-for-ethiopia")
train = ds["train"].to_pandas()
test = ds["test"].to_pandas()

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

---

## Schema

| Column | Type | Null % | Range / Sample Values |
|---|---|---|---|
| `id` | int64 | 0.0% | 7340.0 – 568501.0 (mean 323001.1849) |
| `relid` | object | 0.0% | ETH-2020-1-555-0, ETH-2000-2-193-1.4, ETH-2018-2-5349-0 |
| `year` | int64 | 0.0% | 1989.0 – 2024.0 (mean 2016.1287) |
| `active_year` | bool | 0.0% | |
| `code_status` | object | 0.0% | Clear |
| `type_of_violence` | int64 | 0.0% | 1.0 – 3.0 (mean 1.6678) |
| `conflict_dset_id` | int64 | 0.0% | 91.0 – 17936.0 (mean 2142.3329) |
| `conflict_new_id` | int64 | 0.0% | 267.0 – 16235.0 (mean 2535.7607) |
| `conflict_name` | object | 0.0% | Ethiopia: Oromiya, Government of Ethiopia - Civilians, Ethiopia: Government |
| `dyad_dset_id` | int64 | 0.0% | 91.0 – 17936.0 (mean 5339.3806) |
| `dyad_new_id` | int64 | 0.0% | 555.0 – 17936.0 (mean 5973.41) |
| `dyad_name` | object | 0.0% | Government of Ethiopia - Civilians, Government of Ethiopia - OLA, Government of Ethiopia - TPLF |
| `side_a_dset_id` | int64 | 0.0% | 91.0 – 8803.0 (mean 576.757) |
| `side_a_new_id` | int64 | 0.0% | 91.0 – 8803.0 (mean 576.757) |
| `side_a` | object | 0.0% | Government of Ethiopia, Government of Eritrea, Fano |
| `side_b_dset_id` | int64 | 0.0% | 97.0 – 9999.0 (mean 5158.8682) |
| `side_b_new_id` | int64 | 0.0% | 1.0 – 8196.0 (mean 2196.2228) |
| `side_b` | object | 0.0% | Civilians, OLA, TPLF |
| `number_of_sources` | int64 | 0.0% | -1.0 – 22.0 (mean 0.8517) |
| `source_article` | object | 0.0% | "“SUCH A BRUTAL CRACKDOWN” ,2016-06-30,Annex 1: Documented Killings ", Summary of Oromo Liberation Army (OLA) activities (2012)" at http://www.oromoliberationfront.org/News/2013/Birkii_Beeksisa_Alaaf.pdf, "OLA,2023-12-26,Military Update No: 2023-12-26" |
| `source_office` | object | 27.4% | ONM-ABO, My views on news, BBC Monitoring Africa |
| `source_date` | object | 27.4% | 2016-06-30, 2023-12-26, 2022-10-31 |
| `source_headline` | object | 27.4% | Annex 1: Documented Killings, Military Update No: 2023-12-26, INVESTIGATION AND DOCUMENTATION OF VIOLATIONS OF HUMAN RIGHTS AND HUMANITARIAN LAWS DURING THE WAR ON TIGRAY Preliminary Report of South-eastern Zone |
| `source_original` | object | 35.2% | |
| `where_prec` | int64 | 0.0% | 1.0 – 6.0 (mean 2.7575) |
| `where_coordinates` | object | 0.0% | |
| `where_description` | object | 1.0% | |
| `adm_1` | object | 0.9% | |
| `adm_2` | object | 10.2% | |
| `latitude` | float64 | 0.0% | 3.5366 – 16.1833 (mean 9.5486) |
| `longitude` | float64 | 0.0% | 33.441 – 47.5672 (mean 39.1614) |
| `geom_wkt` | object | 0.0% | |
| `priogrid_gid` | int64 | 0.0% | 135077.0 – 153079.0 (mean 143463.0158) |
| `country` | object | 0.0% | |
| `iso3` | object | 0.0% | |
| `country_id` | int64 | 0.0% | 530.0 – 530.0 (mean 530.0) |
| `region` | object | 0.0% | |
| `event_clarity` | int64 | 0.0% | 1.0 – 2.0 (mean 1.2031) |
| `date_prec` | int64 | 0.0% | 1.0 – 5.0 (mean 1.6738) |
| `date_start` | datetime64[ns] | 0.0% | |
| `date_end` | datetime64[ns] | 0.0% | |
| `deaths_a` | int64 | 0.0% | 0.0 – 121848.0 (mean 62.2802) |
| `deaths_b` | int64 | 0.0% | |
| `deaths_civilians` | int64 | 0.0% | |
| `deaths_unknown` | int64 | 0.0% | |
| `best` | int64 | 0.0% | |
| `high` | int64 | 0.0% | |
| `low` | int64 | 0.0% | |
| `gwnoa` | float64 | 13.5% | |
| `esa_source` | object | 0.0% | |
| `esa_processed` | object | 0.0% | |

---

## Numeric Summary

| Column | Min | Max | Mean | Median |
|---|---|---|---|---|
| `id` | 7340.0 | 568501.0 | 323001.1849 | 424175.0 |
| `year` | 1989.0 | 2024.0 | 2016.1287 | 2021.0 |
| `type_of_violence` | 1.0 | 3.0 | 1.6678 | 1.0 |
| `conflict_dset_id` | 91.0 | 17936.0 | 2142.3329 | 329.0 |
| `conflict_new_id` | 267.0 | 16235.0 | 2535.7607 | 413.0 |
| `dyad_dset_id` | 91.0 | 17936.0 | 5339.3806 | 719.0 |
| `dyad_new_id` | 555.0 | 17936.0 | 5973.41 | 941.0 |
| `side_a_dset_id` | 91.0 | 8803.0 | 576.757 | 97.0 |
| `side_a_new_id` | 91.0 | 8803.0 | 576.757 | 97.0 |
| `side_b_dset_id` | 97.0 | 9999.0 | 5158.8682 | 7595.0 |
| `side_b_new_id` | 1.0 | 8196.0 | 2196.2228 | 497.0 |
| `number_of_sources` | -1.0 | 22.0 | 0.8517 | 1.0 |
| `where_prec` | 1.0 | 6.0 | 2.7575 | 2.0 |
| `latitude` | 3.5366 | 16.1833 | 9.5486 | 9.1728 |
| `longitude` | 33.441 | 47.5672 | 39.1614 | 39.0 |

---

## 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: `gwnob`. 2 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 HDX 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: `source_office`, `source_date`, `source_headline`, `source_original`.
- Refer to the [original HDX dataset page](https://data.humdata.org/dataset/ucdp-data-for-ethiopia) for the publisher's own methodology notes and caveats.

---

## Citation

```bibtex
@dataset{hdx_africa_ucdp_data_for_ethiopia,
title = {Ethiopia - Data on Conflict Events},
author = {HDX},
year = {2026},
url = {https://data.humdata.org/dataset/ucdp-data-for-ethiopia},
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.*