---
annotations_creators:
- no-annotation
language_creators:
- found
language:
- en
license: cc-by-sa-4.0
multilinguality:
- monolingual
size_categories:
- n<1K
source_datasets:
- original
task_categories:
- tabular-classification
- tabular-regression
task_ids: []
tags:
- africa
- humanitarian
- hdx
- electric-sheep-africa
- conflict-violence
- fatalities
- forecasting
- hxl
- mdg
pretty_name: "Madagascar - VIEWS conflict forecasts"
dataset_info:
  splits:
    - name: train
      num_examples: 28
    - name: test
      num_examples: 7
---

# Madagascar - VIEWS conflict forecasts

**Publisher:** Violence & Impacts Early-Warning System · **Source:** [HDX](https://data.humdata.org/dataset/mdg-views-conflict-forecasts) · **License:** `cc-by-sa` · **Updated:** 2026-04-01

---

## Abstract

The Violence & Impacts Early-Warning System (VIEWS) is an award-winning conflict prediction system that generates monthly forecasts for violent conflicts across the world up to three years in advance. It is supported by the iterative research and development activities undertaken by the VIEWS consortium.

Each row in this dataset represents country-level aggregates. Data was last updated on HDX on 2026-04-01. Geographic scope: **MDG**.

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

---

## Dataset Characteristics

| | |
|---|---|
| **Domain** | Conflict and security |
| **Unit of observation** | Country-level aggregates |
| **Rows (total)** | 36 |
| **Columns** | 12 (8 numeric, 4 categorical, 0 datetime) |
| **Train split** | 28 rows |
| **Test split** | 7 rows |
| **Geographic scope** | MDG |
| **Publisher** | Violence & Impacts Early-Warning System |
| **HDX last updated** | 2026-04-01 |

---

## Variables

**Geographic** — `country_id` (range 172.0–172.0), `isoab` (MDG), `year` (range 2026.0–2029.0).

**Temporal** — `month_id` (range 555.0–590.0), `month` (range 1.0–12.0).

**Identifier / Metadata** — `name` (Madagascar), `gwcode` (range 580.0–580.0), `esa_source` (HDX), `esa_processed` (2026-04-05).

**Other** — `main_mean_ln` (range 0.0167–0.0955), `main_mean` (range 0.0169–0.1002), `main_dich` (range 0.0–0.0).

---

## Quick Start

```python
from datasets import load_dataset

ds    = load_dataset("electricsheepafrica/africa-mdg-views-conflict-forecasts")
train = ds["train"].to_pandas()
test  = ds["test"].to_pandas()

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

---

## Schema

| Column | Type | Null % | Range / Sample Values |
|---|---|---|---|
| `country_id` | int64 | 0.0% | 172.0 – 172.0 (mean 172.0) |
| `month_id` | int64 | 0.0% | 555.0 – 590.0 (mean 572.5) |
| `name` | object | 0.0% | Madagascar |
| `gwcode` | int64 | 0.0% | 580.0 – 580.0 (mean 580.0) |
| `isoab` | object | 0.0% | MDG |
| `year` | int64 | 0.0% | 2026.0 – 2029.0 (mean 2027.1667) |
| `month` | int64 | 0.0% | 1.0 – 12.0 (mean 6.5) |
| `main_mean_ln` | float64 | 0.0% | 0.0167 – 0.0955 (mean 0.0608) |
| `main_mean` | float64 | 0.0% | 0.0169 – 0.1002 (mean 0.0629) |
| `main_dich` | float64 | 0.0% | 0.0 – 0.0 (mean 0.0) |
| `esa_source` | object | 0.0% | HDX |
| `esa_processed` | object | 0.0% | 2026-04-05 |

---

## Numeric Summary

| Column | Min | Max | Mean | Median |
|---|---|---|---|---|
| `country_id` | 172.0 | 172.0 | 172.0 | 172.0 |
| `month_id` | 555.0 | 590.0 | 572.5 | 572.5 |
| `gwcode` | 580.0 | 580.0 | 580.0 | 580.0 |
| `year` | 2026.0 | 2029.0 | 2027.1667 | 2027.0 |
| `month` | 1.0 | 12.0 | 6.5 | 6.5 |
| `main_mean_ln` | 0.0167 | 0.0955 | 0.0608 | 0.0609 |
| `main_mean` | 0.0169 | 0.1002 | 0.0629 | 0.0628 |
| `main_dich` | 0.0 | 0.0 | 0.0 | 0.0 |

---

## 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`. 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 Violence & Impacts Early-Warning System 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/mdg-views-conflict-forecasts) for the publisher's own methodology notes and caveats.

---

## Citation

```bibtex
@dataset{hdx_africa_mdg_views_conflict_forecasts,
  title     = {Madagascar - VIEWS conflict forecasts},
  author    = {Violence & Impacts Early-Warning System},
  year      = {2026},
  url       = {https://data.humdata.org/dataset/mdg-views-conflict-forecasts},
  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.*