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.

Knowledge Graph Reasoning Neural Network Based on Dynamic Programming

Neural Axiom Network for Knowledge Graph Reasoning

Dynamic knowledge graph reasoning based on deep reinforcement learning

Neural Probabilistic Logic Learning for Knowledge Graph Reasoning

Adaptive Graph Neural Network with Incremental Learning Mechanism for Knowledge Graph Reasoning

EG-KGR: A Knowledge Graph Reasoning Model Based on Enhanced Graph Sample and Aggregate Inductive Learning Algorithm

Attention-Based Relational Graph Convolutional Network for Knowledge Graph Reasoning

A Review of Knowledge Reasoning Based on Neural Network

Overview of Knowledge Reasoning for Knowledge Graph

Temporal Inductive Path Neural Network for Temporal Knowledge Graph Reasoning

Learning Adaptive Propagation for Knowledge Graph Reasoning

An Overview of Knowledge Graph Reasoning: Key Technologies and Applications

GK: A Graphormer Architecture Based Knowledge Graph Reasoning Model for Link Prediction

Dynamically Pruned Message Passing Networks for Large-Scale Knowledge Graph Reasoning

River of No Return: Graph Percolation Embeddings for Efficient Knowledge Graph Reasoning

Temporal Knowledge Graph Reasoning Based on Graph Convolution Network

Neural, symbolic and neural-symbolic reasoning on knowledge graphs

Hierarchical Hypergraph Recurrent Attention Network for Temporal Knowledge Graph Reasoning

A*Net: A Scalable Path-based Reasoning Approach for Knowledge Graphs

Single Source Path-Based Graph Neural Network for Inductive Knowledge Graph Reasoning

Distance-Based Propagation for Efficient Knowledge Graph Reasoning