Knowledge graph alignment model training method, alignment method, device and equipment

Abstract

The invention discloses a knowledge graph alignment model training method, an alignment method, a device and equipment and belongs to the technical field of artificial intelligence and cloud. The method comprises the following steps: acquiring a first entity pair set, wherein the first entity pair set comprises a plurality of first entity pairs of which alignment results are not labeled; based on the predicted alignment probability of each first entity pair, screening out a plurality of first candidate entity pairs from each first entity pair; calculating the difficult alignment degree of each first candidate entity pair; screening out a plurality of first target entity pairs from each first candidate entity pair based on the difficult alignment degree of each first candidate entity pair; and obtaining a labeling alignment result of each first target entity pair, and obtaining a target knowledge graph alignment model according to the predicted alignment probability and the labeling alignment result of each first target entity pair. According to the method, through two times of screening, the number of entity pairs needing to be labeled is greatly reduced, the time cost is saved, the training speed of the model is increased, and the alignment efficiency is improved.

Description

technical field [0001] The embodiments of the present application relate to the fields of artificial intelligence and cloud technology, and in particular to a training method, alignment method, device and equipment for a knowledge graph alignment model. Background technique [0002] The knowledge graph is composed of interconnected entities and the relationship between entities. The knowledge graph alignment technology is of great significance for the construction of large-scale high-quality knowledge graphs. Usually, based on the trained knowledge graph alignment model, the entities in the two knowledge graphs are aligned, so as to introduce the content of other knowledge graphs into one knowledge graph, thereby integrating knowledge graphs of different granularities and different vertical domains. [0003] In related technologies, two knowledge graphs are obtained in advance, and for any entity in one knowledge graph and any entity in the other knowledge graph, the two ent...

AI Technical Summary

Problems solved by technology

[0004] In the above method, the number of training data sets is very large, and manual labeling of each entity pair in the training data set takes a lot of time, which affects the training speed of the model, thereby affecting the alignment efficiency

Images