GeoVectors-Africa-location (v0.1)

Description

The GeoVectors corpus is a comprehensive large-scale linked open corpus of OpenStreetMap (https://www.openstreetmap.org/) entity embeddings that provides latent representations of over 980 million entities. The GeoVectors capture the semantic and geographic similarities of OpenStreetMap entities and make them directly accessible to machine learning applications. The "-tags" datasets provide embeddings that capture the semantic similarities of OpenStreetMap entities. The "-location" datasets provide the geographic similarities.

Contents

This dataset was derived from an OpenStreetMap snapshot that was taken on November 10, 2020 (© OpenStreetMap contributors).

We provide the GeoVectors in region-specific subsets. This subset contains location-embeddings for the region "Africa" including the following countries:

Algeria

Angola

Benin

Botswana

Burkina-Faso

Burundi

Cameroon

Canary-Islands

Cape-Verde

Central-African-Republic

Chad

Comores

Congo-Brazzaville

Congo-Democratic-Republic

Djibouti

Egypt

Equatorial-Guinea

Eritrea

Ethiopia

Gabon

Ghana

Guinea

Guinea-Bissau

Ivory-Coast

Kenya

Lesotho

Liberia

Libya

Madagascar

Malawi

Mali

Mauritania

Mauritius

Morocco

Mozambique

Namibia

Niger

Nigeria

Rwanda

Saint-Helena-Ascension-and-Tristan-Da-Cunha

Sao-Tome-and-Principe

Senegal-and-Gambia

Seychelles

Sierra-Leone

Somalia

South-Africa

South-Sudan

Sudan

Swaziland

Tanzania

Togo

Tunisia

Uganda

Zambia

Zimbabwe

File format

The embeddings are provided in the tab-separated values (tsv) format. Each row contains the embedding of a single OpenStreetMap entity. The first column contains the OpenStreetMap type and the second column contains the OpenStreetMap id of the respective entity. The type can either be node (n), way (w), or relation (r). The remaining columns represent the dimensions of the embedding space. (See also header.tsv)

Further information:

For further information, please visit http://geovectors.l3s.uni-hannover.de

Funding:

This work was partially funded by DFG, German Research Foundation (“WorldKG", DE 2299/2-1), the Federal Ministry of Education and Research (BMBF), Germany (“Simple-ML", 01IS18054), the Federal Ministry for Economic Affairs and Energy (BMWi), Germany (“d-E-mand", 01ME19009B), and the European Commission (EU H2020, “smashHit", grant-ID 871477).