electricsheepafrica/africa-drc-3-w-national-june-2015

DRC - 3 W National - June 2015

Publisher: OCHA Democratic Republic of the Congo (DRC) · Source: HDX · License: cc-by-igo · Updated: 2024-08-08

---

Abstract

3W of operational projects DRC

Each row in this dataset represents subnational administrative unit observations. Data was last updated on HDX on 2024-08-08. Geographic scope: COD.

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

---

Dataset Characteristics

Domain	Public health
Unit of observation	Subnational administrative unit observations
Rows (total)	1,851
Columns	26 (6 numeric, 20 categorical, 0 datetime)
Train split	1,480 rows
Test split	370 rows
Geographic scope	COD
Publisher	OCHA Democratic Republic of the Congo (DRC)
HDX last updated	2024-08-08

---

Variables

Geographic — agency_acronym (PAM, IEDA Relief, UNHCR), province (Sud-Kivu, Katanga, Nord-Kivu), district (Sud-Kivu, Nord-Kivu, Maniema), territory (Rutshuru, Kalehe, Walikale), location (Toutes, toutes les structures, toutes les localités) and 2 others.

Temporal — start_date, end_date.

Demographic — household (range 1.0–5000000.0).

Identifier / Metadata — adm1pcode (range 50.0–70.0), adm2pcode (range 501.0–707.0), adm3pcode (range 5010.0–7076.0), esa_source, esa_processed.

Other — health_division (Toutes, Pweto, Dungu), titre_de_projet (Suivi et Evaluation des Activités relevant du Mandat du HCR/Monitoring de Protection, Gouvernance pour la protection des enfants, Nutrition), cluster (Protection, Santé, Nutrition), cluster_cleaned (Protection, Securite Alimentaire, Santé), project and 6 others.

---

Quick Start

from datasets import load_dataset

ds    = load_dataset("electricsheepafrica/africa-drc-3-w-national-june-2015")
train = ds["train"].to_pandas()
test  = ds["test"].to_pandas()

print(train.shape)
train.head()

---

Schema

Column	Type	Null %	Range / Sample Values
agency_acronym	object	0.2%	PAM, IEDA Relief, UNHCR
province	object	0.0%	Sud-Kivu, Katanga, Nord-Kivu
adm1pcode	int64	0.0%	50.0 – 70.0 (mean 61.8622)
district	object	0.0%	Sud-Kivu, Nord-Kivu, Maniema
adm2pcode	int64	0.0%	501.0 – 707.0 (mean 618.3798)
territory	object	0.0%	Rutshuru, Kalehe, Walikale
adm3pcode	float64	0.5%	5010.0 – 7076.0 (mean 6193.3286)
health_division	object	18.4%	Toutes, Pweto, Dungu
location	object	27.4%	Toutes, toutes les structures, toutes les localités
titre_de_projet	object	15.1%	Suivi et Evaluation des Activités relevant du Mandat du HCR/Monitoring de Protection, Gouvernance pour la protection des enfants, Nutrition
cluster	object	0.3%	Protection, Santé, Nutrition
cluster_cleaned	object	0.0%	Protection, Securite Alimentaire, Santé
activity_description	object	44.5%	Monitoring de protection, Réhabilitation nutritionnelle, Distribution alimentaire
project	object	18.1%
project_description	object	23.6%
donors	object	21.1%
household	float64	40.3%	1.0 – 5000000.0 (mean 214793.5644)
women	float64	71.0%	2.0 – 642852.0 (mean 113061.825)
men	float64	75.8%	1.0 – 484205.0 (mean 100668.808)
type_of_beneficiary	object	19.1%
implenting_partners	object	46.7%
start_date	object	8.8%
end_date	object	11.3%
notes	object	68.2%
esa_source	object	0.0%
esa_processed	object	0.0%

---

Numeric Summary

Column	Min	Max	Mean	Median
adm1pcode	50.0	70.0	61.8622	62.0
adm2pcode	501.0	707.0	618.3798	622.0
adm3pcode	5010.0	7076.0	6193.3286	6224.0
household	1.0	5000000.0	214793.5644	5737.0
women	2.0	642852.0	113061.825	1923.0
men	1.0	484205.0	100668.808	2000.0

---

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: sub_cluster, sub_type_project. 4 exact duplicate rows were removed. 3 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 OCHA Democratic Republic of the Congo (DRC) 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: location, activity_description, project_description, donors, household, women, men, implenting_partners....
• Refer to the original HDX dataset page for the publisher's own methodology notes and caveats.

---

Citation

@dataset{hdx_africa_drc_3_w_national_june_2015,
  title     = {DRC - 3 W National - June 2015},
  author    = {OCHA Democratic Republic of the Congo (DRC)},
  year      = {2024},
  url       = {https://data.humdata.org/dataset/drc-3-w-national-june-2015},
  note      = {Repackaged for machine learning by Electric Sheep Africa (https://huggingface.co/electricsheepafrica)}
}

---

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