Heterogeneous ontology matching method and system based on graph

Abstract

The invention discloses a heterogeneous ontology matching method and system based on a graph. Based on relational space and spectral radius of the graph, a correct matching pair is acquired by determining the relational space, establishing the graph on the relational space and taking the spectral radius of the graph as a standard selection matching pair, the matching system of heterogeneous ontology based on the graph comprises a relational space module, a graph construction module and a matching module; the graph construction module particularly comprises a text similarity computing submodule, a structural consistency computing submodule and an edge weight value assignment submodule; so that the heterogeneous ontology matching based on the relational space and the spectral radius of the graph is achieved. The heterogeneous ontology matching method and system based on the graph better solves the problem of the heterogeneous ontology matching, the implementation is easy, the precision is high, and the effect is good.

Description

technical field [0001] The invention belongs to the field of information processing, relates to ontology and ontology matching, and in particular to a graph-based heterogeneous ontology matching method and system. Background technique [0002] Ontology in the field of information processing can be regarded as a collection of general and special knowledge resources. Ontology provides rich knowledge and lays a solid foundation for artificial intelligence applications such as information extraction and natural language processing. In recent years, ontology has been increasingly used in other fields to achieve the goal of intelligentizing the system as a whole. Ontology includes the conceptual definition of a knowledge or technical field, and its components are entities and the relationships between entities. Generally speaking, there are three types of entities in ontology, which are class entities, attribute entities and individual entities; the relationship in ontology descr...

AI Technical Summary

Problems solved by technology

However, the ready-made methods and tools for semantic analysis are less accurate and the analysis rate is not high, which limits the accuracy of text matching methods;

[0007] (3) The existing structure matching methods often only use the local structure information of the ontology, such as: the degree of a node, the degree of a node's neighbor nodes, the cluster of the node, etc.

;The lack of global structure information of the ontology leads to a greater impact on the accuracy of the structure matching method;

[0008] (4) The structure matching method is difficult to fuse the text information of the entity in the ontology

The structural matching method regards the ontology as a graph composed of nodes and edges, which makes it difficult for the text information of entities to be naturally introduced into this graph structure

If the textual information of the entities in the ontology is lacking, the accuracy of heterogeneous ontology matching will also be greatly negatively affected by relying only on the structural information of the ontology.

Images