MOSAIC-KG: An Interdisciplinary Knowledge Graph for Explaining Cascading Climate Effects on Health
Authors/Creators
-
Assoudi, Houdhem
(Researcher)1, 2, 3
-
Armant, Vincent
(Supervisor)4
-
Mouakher, Amira
(Supervisor)5
-
Roux, Emmanuel
(Project manager)4
-
Desconnets, Jean-Christophe
(Supervisor)6
-
Marques da Costa, Eduarda
(Researcher)7
-
ABRANTES, PATRICIA
(Researcher)8
-
Morgado, Paulo
(Researcher)9
-
Louro, Ana
(Researcher)10, 11
-
Mose, Victor N.
(Researcher)12
-
1.
UMR Espace-Dev
-
2.
Université de Montpellier
-
3.
IRD, Centre de Montpellier
-
4.
Institut de Recherche pour le Développement
- 5. Université de Perpignan Via Domitia Faculté des Sciences Exactes et Expérimentales
- 6. institut de recherche pour le développement
- 7. Institute of Geography and Spatial Planning, University of Lisbon
- 8. Universidade de Lisboa Instituto de Geografia e Ordenamento do Território
- 9. Instituto de Geografia e Ordenamento do Território da Universidade de Lisboa
- 10. Finisterra - Revista Portuguesa de Geografia
-
11.
University of Lisbon
-
12.
African Conservation Centre
Description
MOSAIC-KG is an interdisciplinary knowledge graph designed to organize and explore the cascading effects of severe climate events on environmental, animal, and human health, with a particular focus on drought impacts on Maasai pastoral communities in East Africa.
The resource combines an expert-curated conceptual model of climate–health impact pathways with knowledge extracted from interdisciplinary scientific literature. It represents scientific assertions, contextual qualifiers (such as mechanisms, conditions, locations, and affected populations), and source references within document-local knowledge graphs.
By connecting general impact pathways to detailed scientific findings and their provenance, MOSAIC-KG supports cross-domain knowledge exploration, evidence retrieval, and the development of explainable retrieval-augmented generation (RAG) applications.
The dataset includes RDF knowledge graphs, conceptual relationships, document-level knowledge representations, and provenance links. It is intended for research in climate–health interactions, One Health, knowledge representation, and information retrieval.