Federal learning-based data matching method and device

Abstract

The invention provides a data matching method and device based on federated learning, and the method comprises the steps: respectively calculating the similarity between a first data label corresponding to a first data set and a plurality of second data labels corresponding to a plurality of second data sets, and generating a plurality of label matching degrees; respectively calculating the similarity between a first data feature set corresponding to the first data set and a plurality of second data feature sets corresponding to the plurality of second data sets, and generating a plurality of data content matching degrees; and based on the label matching degree and the data content matching degree between the first data set and the same second data set, determining a target data set from the plurality of second data sets as federal matching data of the first data set. According to the data matching method based on federated learning, the comprehensiveness, precision and accuracy of a longitudinal federated matching result are remarkably improved, and the subsequent model effect can be improved.

Description

technical field [0001] The present invention relates to the technical field of federated learning, in particular to a data matching method and device based on federated learning. Background technique [0002] When conducting vertical federated learning training, the initiator and data provider must first perform data alignment, and then complete subsequent model training with the participation of the coordinator. Before that, the federated data must be matched. Related matching technologies mainly include matching strategies such as local search, three-party recommendation, and fixed participants. However, the data matched by these methods has problems such as limited data volume, poor data quality, and serious data homogeneity. Contents of the invention [0003] The present invention provides a data matching method and device based on federated learning, which is used to solve the defect of poor data matching effect in the process of longitudinal federated learning in the...

AI Technical Summary

Problems solved by technology

[0003] The present invention provides a data matching method and device based on federated learning, which is used to solve the defect of poor data matching effect in the process of vertical federated learning in the prior art and realize high-quality data matching

Images