Large-scale ontology merging method fusing representation learning and divide-and-conquer strategies

Abstract

The invention discloses a large-scale ontology merging method fusing representation learning and divide-and-conquer strategies, which comprises the following steps of: 1) for two ontologies to be merged, learning semantic representation of composition elements of each ontology in a hypothetical public semantic space; 2) dividing all categories in each ontology into a plurality of disjoint categoryclusters according to the ontology hierarchical structure; according to the hierarchical structure of the categories in the same category cluster in the ontology, recovering the relationship among the categories in the category cluster to obtain a block set of the ontology; 3) generating a block mapping between the two ontologies according to the block sets of the two ontologies to be merged, andaligning the blocks based on the semantic representations of the composition elements of the ontologies; and 4) dividing the aligned ontologies into a source ontology and a target ontology, merging equivalent categories between the source ontology and the target ontology into a common category, putting the common category into a merging ontology, and then putting the rest category information inthe source ontology into the merging ontology to complete merging of the two ontologies.

Description

technical field [0001] The invention belongs to related technologies of network data-oriented knowledge base construction and merging, and in particular relates to a large-scale ontology merging method that integrates representation learning and divide-and-conquer strategies. Background technique [0002] A knowledge base is a collection of interconnected knowledge organized and managed by a certain knowledge representation. Although the definition of knowledge is still a controversial issue in epistemology, in the field of knowledge engineering, the elements of knowledge description generally include classification, entity, relationship, attribute and other elements. Ontology refers to a formalized, clear and detailed description of the shared concept system, which includes elements such as categories, category attributes, and relationships between categories, and is used to semantically group or organize knowledge items in the knowledge base. Semantic annotation. The mai...

AI Technical Summary

Problems solved by technology

However, this strategy is entirely dependent on lexical representations, making it difficult to distinguish between synonymous and polysemy cases

[0004] (2) With the help of external dictionaries or ontology background knowledge strategies, the context information of ontology elements is enriched through external information, but this strategy is limited by the coverage of dictionaries or the richness of background knowledge

The greedy-based method may be an effective method for dealing with large-scale ontology merging tasks, but due to its greedy nature, it is difficult to correct previous errors when merging decisions, resulting in the method not being able to guarantee that the two ontologies are globally optimal The merged result of

[0008] To sum up, at present, there is still a lack of effective methods for the problem of large-scale ontology merging, especially on the premise of ensuring the accuracy of ontology merging, reducing the execution time of large-scale ontology merging, and adapting to the feasibility of large-scale ontology merging. scalability requirements

Images