LLMs4OL 2025: Large Language Models for Ontology Learning
The 2nd LLMs4OL Challenge @ ISWC 2025
ISWC 2025, Nara, Japan | 2-6 November
Embedded Files
Task C - Taxonomy Discovery
Task C - Taxonomy Discovery
Discover the taxonomic hierarchy between type pairs.
Discover the taxonomic hierarchy between type pairs.
Taxonomy discovery focuses on identifying hierarchical relationships between types, enabling the construction of taxonomic structures (i.e., is-a relationships). Given a pair of terms or types, the task determines whether one is a subclass of the other. For example, discovering that "Sedan" is a subclass of "Car" contributes to structuring domain knowledge in a way that supports reasoning and inferencing in ontology-driven applications.
SubTasks
SubTasks
Given a set of ontological types, identify type pairs that form a taxonomic (is-a) relationship.
- SubTask C1 – OBI
Focuses on biomedical investigation types, is-a relationships using the Ontology for Biomedical Investigations (OBI).
- SubTask C2 – MatOnto
This subtask covers materials science taxonomies using the Material Ontology.
- SubTask C3 – SWEET
Uses the SWEET Ontology to uncover hierarchical structures in Earth and environmental science.
- SubTask C4 – DOID
Centers on discovering taxonomic relationships between medical disease types.
- SubTask C5 – SchemaOrg
Focuses on general-purpose web knowledge concepts using the widely adopted Schema.org vocabulary.
- SubTask C6 – PROCO
Involves hierarchical reasoning over chemical process-related entities.
- SubTask C7 – FoodOn
Discovers food-related class hierarchies, such as food types, ingredients, and preparation forms.
- SubTask C8 – PO
Classifies hierarchical plant biology concepts such as plant structures, stages, and growth patterns.
Datasets
Datasets
The datasets per given SubTasks C1, C2, ..., and C8 are available at the "TaskC-TaxonomyDiscovery" directory of the challenge repository here https://github.com/sciknoworg/LLMs4OL-Challenge/tree/main/2025.
Per dataset, two key files were given as follows:
The file with "_types.txt" suffix is a plain text file listing types (one per line). This is the list of types for forming taxonomic (is-a) relations.
A JSON file with "_pairs.json" suffix that represents "is-a" relationships between two types. This file contains ground truth with the following format.
[
{
"ID": "TR_515d09c1",
"parent": "distillation",
"child": "simple distillation"
},
....
]
🧪 During the testing phase, only a list of types will be provided. Participants are expected to submit a "_pairs.json" file that represents taxonomic relationships, where each pair consists of a parent and a child.
Page updated
Google Sites
Report abuse