electricsheepasia/asia-displacement-bangladesh-iom-dtm-from-api

Bangladesh IOM Displacement Tracking Matrix (DTM) from API

Publisher: International Organization for Migration (IOM) · Source: HDX · License: hdx-other · Updated: 2026-05-04

---

Abstract

This dataset comes from the International Organization for Migration (IOM)'s displacement tracking matrix (DTM) publicly accessible API. This API allows the humanitarian community, academia, media, government, and non-governmental organizations to utilize the data collected by DTM. The DTM API only provides non-sensitive IDP figures, aggregated at the country, Admin 1 (states, provinces, or equivalent), and Admin 2 (smaller subnational administrative areas) levels. For more detailed information, please see the country-specific DTM datasets on HDX.

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

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

---

Dataset Characteristics

Domain	Conflict and security
Unit of observation	Subnational administrative unit observations
Rows (total)	73
Columns	21 (6 numeric, 14 categorical, 1 datetime)
Train split	58 rows
Test split	14 rows
Geographic scope	BGD
Publisher	International Organization for Migration (IOM)
HDX last updated	2026-05-04

---

Variables

Geographic — admin0name (Bangladesh), admin0pcode (BGD), admin1name (Dhaka, Chittagong, Khulna), admin1pcode (BD30, BD20, BD40), admin2name (Rajbari, Rangpur, Kurigram) and 7 others.

Temporal — reportingdate, monthreportingdate (range 10.0–10.0).

Outcome / Measurement — roundnumber (range 1.0–1.0).

Identifier / Metadata — id (range 360.0–136356.0), numpresentidpind (range 1121.0–4955527.0), esa_source, esa_processed.

Other — operation (Mobility Monitoring due to Disasters), operationstatus.

---

Quick Start

from datasets import load_dataset

ds    = load_dataset("electricsheepafrica/asia-displacement-bangladesh-iom-dtm-from-api")
train = ds["train"].to_pandas()
test  = ds["test"].to_pandas()

print(train.shape)
train.head()

---

Schema

Column	Type	Null %	Range / Sample Values
id	float64	1.4%	360.0 – 136356.0 (mean 59232.5)
operation	object	0.0%	Mobility Monitoring due to Disasters
admin0name	object	0.0%	Bangladesh
admin0pcode	object	0.0%	BGD
admin1name	object	1.4%	Dhaka, Chittagong, Khulna
admin1pcode	object	1.4%	BD30, BD20, BD40
admin2name	object	12.3%	Rajbari, Rangpur, Kurigram
admin2pcode	object	12.3%	BD3082, BD5585, BD5549
adminlevel	int64	0.0%	0.0 – 2.0 (mean 1.863)
numpresentidpind	int64	0.0%	1121.0 – 4955527.0 (mean 203651.7945)
reportingdate	datetime64[ns]	0.0%
yearreportingdate	int64	0.0%	2025.0 – 2025.0 (mean 2025.0)
monthreportingdate	int64	0.0%	10.0 – 10.0 (mean 10.0)
roundnumber	int64	0.0%	1.0 – 1.0 (mean 1.0)
displacementreason	object	0.0%	Natural disaster
idporiginadmin1name	object	0.0%	Not available
idporiginadmin1pcode	object	0.0%	Not available
assessmenttype	object	0.0%
operationstatus	object	0.0%
esa_source	object	0.0%
esa_processed	object	0.0%

---

Numeric Summary

Column	Min	Max	Mean	Median
id	360.0	136356.0	59232.5	56194.0
adminlevel	0.0	2.0	1.863	2.0
numpresentidpind	1121.0	4955527.0	203651.7945	60003.0
yearreportingdate	2025.0	2025.0	2025.0	2025.0
monthreportingdate	10.0	10.0	10.0	10.0
roundnumber	1.0	1.0	1.0	1.0

---

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: numbermales, numberfemales. 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 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_asia_displacement_bangladesh_iom_dtm_from_api,
  title     = {Bangladesh IOM Displacement Tracking Matrix (DTM) from API},
  author    = {International Organization for Migration (IOM)},
  year      = {2026},
  url       = {https://data.humdata.org/dataset/bgd-iom-dtm-from-api},
  note      = {Repackaged for machine learning by Electric Sheep Africa (https://huggingface.co/electricsheepafrica)}
}

---

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