electricsheepafrica/africa-drought-somalia

Somalia Displacement - [IDPs] - Emergency Trends Tracking (ETT) - Bay Region - [IOM DTM]

Publisher: International Organization for Migration (IOM) · Source: HDX · License: hdx-other · Updated: 2024-03-07

---

Abstract

Emergency Trends Tracking (ETT) is a crisis-based tool that tracks sudden displacement triggered by specific events or emerging crises. The objective of ETT is to help prioritize humanitarian response and to enable partners to deliver rapid assistance. Based on previous drought induced displacement patterns, and the ones observed since the beginning of the drought, the humanitarian community expects that people will move from rural to urban areas in search of humanitarian services. Consequently, this ETT tool which concentrates only on drought induced displacements, focuses on the main urban centers and surrounding villages for each district. The data is collected through Key Informant Interviews (KIIs) at the location level, from Sunday to Wednesday every week. All locations assessed are monitored each week.

Each row in this dataset represents subnational administrative unit observations. Temporal coverage is indicated by the data_collection column(s). Geographic scope: SOM.

Curated into ML-ready Parquet format by Electric Sheep Africa.

---

Dataset Characteristics

Domain	Forced displacement and migration
Unit of observation	Subnational administrative unit observations
Rows (total)	48
Columns	11 (1 numeric, 9 categorical, 1 datetime)
Train split	38 rows
Test split	9 rows
Geographic scope	SOM
Publisher	International Organization for Migration (IOM)
HDX last updated	2024-03-07

---

Variables

Geographic — region (Bay, #adm1+name), district (Baidoa, #adm2+name), location (#adm3+name, Buurow, Yaaq3_2).

Demographic — cccm_nat_coverage (Yes), dtm_ett_coverage (No).

Identifier / Metadata — esa_source (HDX), esa_processed (2026-04-24).

Other — round (NAT_33), data_collection, new_arrivals (range 2.0–62.0), neighborhood (Berdale, Holwadag, Isha).

---

Quick Start

from datasets import load_dataset

ds    = load_dataset("electricsheepafrica/africa-drought-somalia")
train = ds["train"].to_pandas()
test  = ds["test"].to_pandas()

print(train.shape)
train.head()

---

Schema

Column	Type	Null %	Range / Sample Values
round	object	2.1%	NAT_33
data_collection	datetime64[ns]	2.1%
region	object	0.0%	Bay, #adm1+name
district	object	0.0%	Baidoa, #adm2+name
location	object	0.0%	#adm3+name, Buurow, Yaaq3_2
new_arrivals	float64	2.1%	2.0 – 62.0 (mean 18.9362)
neighborhood	object	2.1%	Berdale, Holwadag, Isha
cccm_nat_coverage	object	2.1%	Yes
dtm_ett_coverage	object	2.1%	No
esa_source	object	0.0%	HDX
esa_processed	object	0.0%	2026-04-24

---

Numeric Summary

Column	Min	Max	Mean	Median
new_arrivals	2.0	62.0	18.9362	15.0

---

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. 2 column(s) with >80% missing values were removed: idp_stock_dtm, idp_site. 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 International Organization for Migration (IOM) 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 for the publisher's own methodology notes and caveats.

---

Citation

@dataset{hdx_africa_drought_somalia,
  title     = {Somalia Displacement - [IDPs] -  Emergency Trends Tracking (ETT) - Bay Region - [IOM DTM]},
  author    = {International Organization for Migration (IOM)},
  year      = {2024},
  url       = {https://data.humdata.org/dataset/somalia-displacement-idps-emergency-trends-tracking-ett-iom-dtm},
  note      = {Repackaged for machine learning by Electric Sheep Africa (https://huggingface.co/electricsheepafrica)}
}

---

Electric Sheep Africa — Africa's ML dataset infrastructure. Lagos, Nigeria.