electricsheepafrica/africa-niger-most-likely-fewsnet-fipe

---
annotations_creators:
- no-annotation
language_creators:
- found
language:
- en
license: cc-by-4.0
multilinguality:
- monolingual
size_categories:
- n<1K
source_datasets:
- original
task_categories:
- tabular-classification
- other
task_ids: []
tags:
- africa
- humanitarian
- hdx
- electric-sheep-africa
- food-security
- ner
pretty_name: "Niger Most Likely FEWS NET Acutely Food Insecure Population Estimates Data"
dataset_info:
splits:
- name: train
num_examples: 60
- name: test
num_examples: 15
---

# Niger Most Likely FEWS NET Acutely Food Insecure Population Estimates Data

**Publisher:** FEWS NET · **Source:** [HDX](https://data.humdata.org/dataset/niger_most_likely_fewsnet_fipe) · **License:** `cc-by` · **Updated:** 2026-04-01

---

## Abstract

Niger Most Likely FEWS NET Acutely Food Insecure Population Estimates Data from 2019

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

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

---

## Dataset Characteristics

| | |
|---|---|
| **Domain** | Food security and nutrition |
| **Unit of observation** | First-level administrative unit observations |
| **Rows (total)** | 75 |
| **Columns** | 44 (10 numeric, 27 categorical, 7 datetime) |
| **Train split** | 60 rows |
| **Test split** | 15 rows |
| **Geographic scope** | NER |
| **Publisher** | FEWS NET |
| **HDX last updated** | 2026-04-01 |

---

## Variables

**Geographic** — `country` (Niger), `country_code` (NE), `fewsnet_region` (West Africa), `admin_0` (Niger), `specialization_type` and 3 others.

**Temporal** — `datacollectionperiod` (range 310323.0–373073.0), `reporting_date`.

**Outcome / Measurement** — `phase`, `low_value` (range 100000.0–2500000.0), `high_value` (range 499999.0–4999999.0), `value` (range 100000.0–2500000.0), `phase_name`.

**Identifier / Metadata** — `source_organization` (FEWS NET), `source_document` (Food Assistance Outlook Brief), `geographic_unit_full_name` (Niger), `geographic_unit_name` (Niger), `fnid` (NE) and 8 others.

**Other** — `geographic_group` (Western Africa), `indicator_abbreviation`, `projection_start`, `projection_end`, `status` and 11 others.

---

## Quick Start

```python
from datasets import load_dataset

ds = load_dataset("electricsheepafrica/africa-niger-most-likely-fewsnet-fipe")
train = ds["train"].to_pandas()
test = ds["test"].to_pandas()

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

---

## Schema

| Column | Type | Null % | Range / Sample Values |
|---|---|---|---|
| `source_organization` | object | 0.0% | FEWS NET |
| `source_document` | object | 0.0% | Food Assistance Outlook Brief |
| `country` | object | 0.0% | Niger |
| `country_code` | object | 0.0% | NE |
| `geographic_group` | object | 0.0% | Western Africa |
| `fewsnet_region` | object | 0.0% | West Africa |
| `geographic_unit_full_name` | object | 0.0% | Niger |
| `geographic_unit_name` | object | 0.0% | Niger |
| `fnid` | object | 0.0% | NE |
| `admin_0` | object | 0.0% | Niger |
| `phase` | object | 0.0% | |
| `scenario_name` | object | 0.0% | |
| `indicator_name` | object | 0.0% | |
| `indicator_abbreviation` | object | 0.0% | |
| `projection_start` | datetime64[ns] | 0.0% | |
| `projection_end` | datetime64[ns] | 0.0% | |
| `status` | object | 0.0% | |
| `low_value` | float64 | 0.0% | 100000.0 – 2500000.0 (mean 1137333.3333) |
| `high_value` | float64 | 0.0% | 499999.0 – 4999999.0 (mean 2193332.3333) |
| `value` | float64 | 0.0% | 100000.0 – 2500000.0 (mean 1137333.3333) |
| `id` | int64 | 0.0% | 33126770.0 – 40657271.0 (mean 34149634.1067) |
| `datacollectionperiod` | int64 | 0.0% | 310323.0 – 373073.0 (mean 319280.3333) |
| `datacollection` | int64 | 0.0% | 325936.0 – 383437.0 (mean 333815.6) |
| `scenario` | object | 0.0% | |
| `geographic_unit` | int64 | 0.0% | 8132.0 – 8132.0 (mean 8132.0) |
| `datasourceorganization` | int64 | 0.0% | 1.0 – 1.0 (mean 1.0) |
| `datasourcedocument` | int64 | 0.0% | 6986.0 – 6986.0 (mean 6986.0) |
| `dataseries` | int64 | 0.0% | 6932815.0 – 6932815.0 (mean 6932815.0) |
| `dataseries_name` | object | 0.0% | |
| `specialization_type` | object | 0.0% | |
| `dataseries_specialization_type` | object | 0.0% | |
| `data_usage_policy` | object | 0.0% | |
| `created` | datetime64[ns] | 0.0% | |
| `modified` | datetime64[ns] | 0.0% | |
| `status_changed` | datetime64[ns] | 0.0% | |
| `collection_status` | object | 0.0% | |
| `collection_status_changed` | datetime64[ns] | 0.0% | |
| `collection_schedule` | object | 0.0% | |
| `reporting_date` | datetime64[ns] | 0.0% | |
| `phase_name` | object | 0.0% | |
| `population_range` | object | 0.0% | |
| `description` | object | 0.0% | |
| `esa_source` | object | 0.0% | |
| `esa_processed` | object | 0.0% | |

---

## Numeric Summary

| Column | Min | Max | Mean | Median |
|---|---|---|---|---|
| `low_value` | 100000.0 | 2500000.0 | 1137333.3333 | 1000000.0 |
| `high_value` | 499999.0 | 4999999.0 | 2193332.3333 | 2499999.0 |
| `value` | 100000.0 | 2500000.0 | 1137333.3333 | 1000000.0 |
| `id` | 33126770.0 | 40657271.0 | 34149634.1067 | 33128882.0 |
| `datacollectionperiod` | 310323.0 | 373073.0 | 319280.3333 | 310397.0 |
| `datacollection` | 325936.0 | 383437.0 | 333815.6 | 325973.0 |
| `geographic_unit` | 8132.0 | 8132.0 | 8132.0 | 8132.0 |
| `datasourceorganization` | 1.0 | 1.0 | 1.0 | 1.0 |
| `datasourcedocument` | 6986.0 | 6986.0 | 6986.0 | 6986.0 |
| `dataseries` | 6932815.0 | 6932815.0 | 6932815.0 | 6932815.0 |

---

## 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`. 7 column(s) with >80% missing values were removed: `admin_1`, `admin_2`, `admin_3`, `admin_4`, `pct_phase3`, `pct_phase4`.... 7 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 FEWS NET 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](https://data.humdata.org/dataset/niger_most_likely_fewsnet_fipe) for the publisher's own methodology notes and caveats.

---

## Citation

```bibtex
@dataset{hdx_africa_niger_most_likely_fewsnet_fipe,
title = {Niger Most Likely FEWS NET Acutely Food Insecure Population Estimates Data},
author = {FEWS NET},
year = {2026},
url = {https://data.humdata.org/dataset/niger_most_likely_fewsnet_fipe},
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.*

annotations_creators:
- 无标注
language_creators:
- 公开获取
language:
- 英语
license: CC BY 4.0
multilinguality:
- 单语言
size_categories:
- 样本数<1000
source_datasets:
- 原始数据集
task_categories:
- 表格分类
- 其他
task_ids: []
tags:
- 非洲
- 人道主义
- HDX
- electric-sheep-africa
- 粮食安全
- 命名实体识别
pretty_name: "尼日尔最可能FEWS NET（Famine Early Warning Systems Network）急性粮食不安全人口估计数据"
dataset_info:
splits:
- name: train
num_examples: 60
- name: test
num_examples: 15

---

# 尼日尔最可能FEWS NET（Famine Early Warning Systems Network）急性粮食不安全人口估计数据

**发布方：** FEWS NET · **来源：** [HDX（Humanitarian Data Exchange，人道主义数据交换）](https://data.humdata.org/dataset/niger_most_likely_fewsnet_fipe) · **许可协议：** `cc-by` · **更新时间：** 2026-04-01

---

## 摘要

本数据集为2019年尼日尔最可能FEWS NET急性粮食不安全人口估计数据。数据集中每一行代表一级行政单元的观测结果，时间覆盖范围由`projection_start`、`projection_end`列标注。地理覆盖范围：**NER（尼日尔国家代码）**。

*本数据集已由[Electric Sheep Africa](https://huggingface.co/electricsheepafrica)整理为适配机器学习的Parquet格式。*

---

## 数据集特征

| | |
|---|---|
| **领域** | 粮食安全与营养 |
| **观测单元** | 一级行政单元观测数据 |
| **总行数** | 75 |
| **列数** | 44（10个数值型、27个分类型、7个日期时间型） |
| **训练集划分** | 60行 |
| **测试集划分** | 15行 |
| **地理覆盖范围** | NER（尼日尔） |
| **发布方** | FEWS NET |
| **HDX最后更新时间** | 2026-04-01 |

---

## 变量分类

**地理类变量** — `country`（尼日尔）、`country_code`（NE）、`fewsnet_region`（西非）、`admin_0`（尼日尔）、`specialization_type`及另外3个变量。
**时间类变量** — `datacollectionperiod`（取值范围310323.0–373073.0）、`reporting_date`。
**结果/测量类变量** — `phase`、`low_value`（取值范围100000.0–2500000.0）、`high_value`（取值范围499999.0–4999999.0）、`value`（取值范围100000.0–2500000.0）、`phase_name`。
**标识符/元数据类变量** — `source_organization`（FEWS NET）、`source_document`（《粮食援助展望简报》）、`geographic_unit_full_name`（尼日尔）、`geographic_unit_name`（尼日尔）、`fnid`（NE）及另外8个变量。
**其他类变量** — `geographic_group`（西非）、`indicator_abbreviation`、`projection_start`、`projection_end`、`status`及另外11个变量。

---

## 快速入门

python
from datasets import load_dataset

ds = load_dataset("electricsheepafrica/africa-niger-most-likely-fewsnet-fipe")
train = ds["train"].to_pandas()
test = ds["test"].to_pandas()

print(train.shape)
train.head()

---

## 数据结构（Schema）

| 列名 | 数据类型 | 空值占比 | 取值范围/示例值 |
|---|---|---|---|
| `source_organization` | object | 0.0% | FEWS NET |
| `source_document` | object | 0.0% | Food Assistance Outlook Brief |
| `country` | object | 0.0% | 尼日尔 |
| `country_code` | object | 0.0% | NE |
| `geographic_group` | object | 0.0% | 西非 |
| `fewsnet_region` | object | 0.0% | 西非 |
| `geographic_unit_full_name` | object | 0.0% | 尼日尔 |
| `geographic_unit_name` | object | 0.0% | 尼日尔 |
| `fnid` | object | 0.0% | NE |
| `admin_0` | object | 0.0% | 尼日尔 |
| `phase` | object | 0.0% | 无 |
| `scenario_name` | object | 0.0% | 无 |
| `indicator_name` | object | 0.0% | 无 |
| `indicator_abbreviation` | object | 0.0% | 无 |
| `projection_start` | datetime64[ns] | 0.0% | 无 |
| `projection_end` | datetime64[ns] | 0.0% | 无 |
| `status` | object | 0.0% | 无 |
| `low_value` | float64 | 0.0% | 100000.0 – 2500000.0（均值1137333.3333） |
| `high_value` | float64 | 0.0% | 499999.0 – 4999999.0（均值2193332.3333） |
| `value` | float64 | 0.0% | 100000.0 – 2500000.0（均值1137333.3333） |
| `id` | int64 | 0.0% | 33126770.0 – 40657271.0（均值34149634.1067） |
| `datacollectionperiod` | int64 | 0.0% | 310323.0 – 373073.0（均值319280.3333） |
| `datacollection` | int64 | 0.0% | 325936.0 – 383437.0（均值333815.6） |
| `scenario` | object | 0.0% | 无 |
| `geographic_unit` | int64 | 0.0% | 8132.0 – 8132.0（均值8132.0） |
| `datasourceorganization` | int64 | 0.0% | 1.0 – 1.0（均值1.0） |
| `datasourcedocument` | int64 | 0.0% | 6986.0 – 6986.0（均值6986.0） |
| `dataseries` | int64 | 0.0% | 6932815.0 – 6932815.0（均值6932815.0） |
| `dataseries_name` | object | 0.0% | 无 |
| `specialization_type` | object | 0.0% | 无 |
| `dataseries_specialization_type` | object | 0.0% | 无 |
| `data_usage_policy` | object | 0.0% | 无 |
| `created` | datetime64[ns] | 0.0% | 无 |
| `modified` | datetime64[ns] | 0.0% | 无 |
| `status_changed` | datetime64[ns] | 0.0% | 无 |
| `collection_status` | object | 0.0% | 无 |
| `collection_status_changed` | datetime64[ns] | 0.0% | 无 |
| `collection_schedule` | object | 0.0% | 无 |
| `reporting_date` | datetime64[ns] | 0.0% | 无 |
| `phase_name` | object | 0.0% | 无 |
| `population_range` | object | 0.0% | 无 |
| `description` | object | 0.0% | 无 |
| `esa_source` | object | 0.0% | 无 |
| `esa_processed` | object | 0.0% | 无 |

---

## 数值型变量统计摘要

| 列名 | 最小值 | 最大值 | 均值 | 中位数 |
|---|---|---|---|---|
| `low_value` | 100000.0 | 2500000.0 | 1137333.3333 | 1000000.0 |
| `high_value` | 499999.0 | 4999999.0 | 2193332.3333 | 2499999.0 |
| `value` | 100000.0 | 2500000.0 | 1137333.3333 | 1000000.0 |
| `id` | 33126770.0 | 40657271.0 | 34149634.1067 | 33128882.0 |
| `datacollectionperiod` | 310323.0 | 373073.0 | 319280.3333 | 310397.0 |
| `datacollection` | 325936.0 | 383437.0 | 333815.6 | 325973.0 |
| `geographic_unit` | 8132.0 | 8132.0 | 8132.0 | 8132.0 |
| `datasourceorganization` | 1.0 | 1.0 | 1.0 | 1.0 |
| `datasourcedocument` | 6986.0 | 6986.0 | 6986.0 | 6986.0 |
| `dataseries` | 6932815.0 | 6932815.0 | 6932815.0 | 6932815.0 |

---

## 数据整理流程

原始数据通过CKAN API从HDX下载，并转换为Parquet格式。列名统一转换为小写并规范为蛇形命名法。将常见缺失值标记（`N/A`、`null`、`none`、`-`、`unknown`、`no data`、`#N/A`）统一替换为`NaN`。移除了7个缺失值占比超过80%的列：`admin_1`、`admin_2`、`admin_3`、`admin_4`、`pct_phase3`、`pct_phase4`等。基于解析成功率（阈值85%），将7列从字符串类型转换为数值型或日期时间型。采用固定随机种子（42）将数据集按80/20划分为训练集与测试集，并保存为Snappy压缩的Parquet格式。

---

## 数据集局限性

- 数据源自FEWS NET，未经过Electric Sheep Africa的独立验证。
- 自动化清洗无法修正原始数据收集中的错报值、定义不一致或抽样偏差问题。
- 请参阅[原始HDX数据集页面](https://data.humdata.org/dataset/niger_most_likely_fewsnet_fipe)获取发布方提供的方法说明与注意事项。

---

## 引用格式

bibtex
@dataset{hdx_africa_niger_most_likely_fewsnet_fipe,
title = {尼日尔最可能FEWS NET急性粮食不安全人口估计数据},
author = {FEWS NET},
year = {2026},
url = {https://data.humdata.org/dataset/niger_most_likely_fewsnet_fipe},
note = {由Electric Sheep Africa（https://huggingface.co/electricsheepafrica）重新打包以适配机器学习场景}
}

---

*[Electric Sheep Africa](https://huggingface.co/electricsheepafrica) — 非洲的机器学习数据集基础设施。尼日利亚拉各斯。*