Embedding Ontologies via Incorporating Extensional and Intensional Knowledge

2024-01-20

arXiv.org

AI Summary

Existing ontology embedding methods struggle to simultaneously capture extensional knowledge (instance-to-concept membership) and intensional knowledge (semantic attributes of concepts). To address this, we propose EIKE, a dual-space unified framework that, for the first time, decouples modeling of the extensional space (instance–concept affiliation) from the intensional space (semantic relations among concepts). EIKE innovatively integrates geometric embedding—specifically a TransE variant—with a pretrained language model (BERT) to jointly represent instances, concepts, and relations. Through multi-space alignment and joint optimization, EIKE achieves significant improvements over state-of-the-art methods on both triple classification and link prediction tasks across three standard benchmarks. Experimental results demonstrate its superior representational capacity and comprehensive modeling capability, validating its effectiveness in unifying extensional and intensional semantics within a single coherent framework.

Technology Category

• Knowledge Representation and Reasoning: Ontologies
• Data Mining & Knowledge Management: Linked Open Data, Knowledge Graphs & KB Completion
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Ontologies contain rich knowledge within domain, which can be divided into two categories, namely extensional knowledge and intensional knowledge. Extensional knowledge provides information about the concrete instances that belong to specific concepts in the ontology, while intensional knowledge details inherent properties, characteristics, and semantic associations among concepts. However, existing ontology embedding approaches fail to take both extensional knowledge and intensional knowledge into fine consideration simultaneously. In this paper, we propose a novel ontology embedding approach named EIKE (Extensional and Intensional Knowledge Embedding) by representing ontologies in two spaces, called extensional space and intensional space. EIKE presents a unified framework for embedding instances, concepts and their relations in an ontology, applying a geometry-based method to model extensional knowledge and a pretrained language model to model intensional knowledge, which can capture both structure information and textual information. Experimental results show that EIKE significantly outperforms state-of-the-art methods in three datasets for both triple classification and link prediction, indicating that EIKE provides a more comprehensive and representative perspective of the domain.

Problem

• Embedding ontologies by combining extensional and intensional knowledge
• Addressing limitations of existing single-perspective ontology representation methods
• Creating unified framework for instances, concepts and relations embedding

Innovation

• Dual-space embedding for extensional and intensional knowledge
• Geometry-based modeling for extensional knowledge representation
• Pretrained language model for intensional knowledge processing