Abstract
Reasoning over knowledge bases such as Semantic Web ontologies enables the discovery of new facts from existing knowledge. Knowledge-enhanced machine learning has motivated the development of neuro-symbolic reasoners, which enable faster but approximate computation of new facts or entailments. Neuro-symbolic methods generate vector representations (embeddings) of entities in a knowledge base. We analyze some ontology embedding methods, by implementing them as neuro-symbolic reasoners and evaluating their predictive performance on the datasets and tasks provided by the Semantic Reasoning Evaluation Challenge 2023. We explore two types of embedding methods: graph-based and model-theoretic. Regarding graph-based embeddings, we evaluated the impact of different combinations of graph representation of ontologies with knowledge graph embedding methods. For model-theoretic embeddings, which create models for theories, we evaluate the impact of using several models, enabling approximate semantic entailment.
Original language | English (US) |
---|---|
State | Published - 2023 |
Event | 1st Scholarly QALD Challenge 2023 and 4th SeMantic Answer Type, Relation and Entity Prediction Tasks Challenge, Scholarly QALD 2023 and SemREC 2023 - Athens, Greece Duration: Nov 6 2023 → Nov 10 2023 |
Conference
Conference | 1st Scholarly QALD Challenge 2023 and 4th SeMantic Answer Type, Relation and Entity Prediction Tasks Challenge, Scholarly QALD 2023 and SemREC 2023 |
---|---|
Country/Territory | Greece |
City | Athens |
Period | 11/6/23 → 11/10/23 |
Keywords
- approximate entailment
- neuro-symbolic AI
- ontology embedding
ASJC Scopus subject areas
- General Computer Science