Abstract:Knowledge graph reasoning (KGR) aims to infer new knowledge or detect noises, which is essential for improving the quality of knowledge graphs. Recently, various KGR techniques, such as symbolic- and embedding-based methods, have been proposed and shown strong reasoning ability. Symbolic-based reasoning methods infer missing triples according to predefined rules or ontologies. Although rules and axioms have proven effective, it is difficult to obtain them. Embedding-based reasoning methods represent entities and relations as vectors, and complete KGs via vector computation. However, they mainly rely on structural information and ignore implicit axiom information not predefined in KGs but can be reflected in data. That is, each correct triple is also a logically consistent triple and satisfies all axioms. In this paper, we propose a novel NeuRal Axiom Network (NeuRAN) framework that combines explicit structural and implicit axiom information without introducing additional ontologies. Specifically, the framework consists of a KG embedding module that preserves the semantics of triples and five axiom modules that encode five kinds of implicit axioms. These axioms correspond to five typical object property expression axioms defined in OWL2, including ObjectPropertyDomain, ObjectPropertyRange, DisjointObjectProperties, IrreflexiveObjectProperty and AsymmetricObjectProperty. The KG embedding module and axiom modules compute the scores that the triple conforms to the semantics and the corresponding axioms, respectively. Compared with KG embedding models and CKRL, our method achieves comparable performance on noise detection and triple classification and achieves significant performance on link prediction. Compared with TransE and TransH, our method improves the link prediction performance on the Hits@1 metric by 22.0% and 20.8% on WN18RR-10% dataset, respectively.

A Kindred Automatic Reasoning Model Based on Large-Scale Chinese Kinship Knowledge Base

The semantic representation and automatic reasoning for Chinese kinship relations

Logic Meaning of Kinship Relations and Kindred Automatic Reasoning

A Situation-based Semantic Scheme for Chinese Kinship Terms

Product Knowledge Reasoning

A Knowledge Representation Method Based on Two-Layer Modeling for Constructing Medical Knowledge Base

The expressive language ALCNHR+K(D) for knowledge reasoning

Tri-Subject Kinship Verification: Understanding the Core of A Family

Graph-based Kinship Reasoning Network

Neural Axiom Network for Knowledge Graph Reasoning

Explore Modeling Relation Information and Direction Information in KBQA

Reasoning Graph Networks for Kinship Verification: From Star-Shaped to Hierarchical

An Extendible Realism-Based Ontology for Kinship

Chain-of-Knowledge: Integrating Knowledge Reasoning into Large Language Models by Learning from Knowledge Graphs

Modeling of complex internal logic for knowledge base completion

Neural-Symbolic Commonsense Reasoner with Relation Predictors

Deep Reasoning with Knowledge Graph for Social Relationship Understanding

Commonsense Generative Model for Chinese Automatic Knowledge Graph Construction

Knowledge-Based Reasoning Network for Relation Detection

Kinship Measurement on Salient Facial Features

MACHINE REASONING BASED ON KNOWLEDGE GRAPH