Temporal feature knowledge graph logic query answering method, medium and equipment

By dynamically adjusting the weights of the logical system and constructing an MLP in the temporal feature knowledge graph, the problem that fixed logical strategies cannot adapt to differences in semantic relevance is solved, improving the accuracy and generalization ability of logical queries, especially when dealing with heterogeneous logical structures.

CN121979979APending Publication Date: 2026-05-05CHINA UNIV OF GEOSCIENCES (WUHAN) +1
0 Cites 0 Cited by

Patent Information

Application Number
CN202512016434.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-12-30
Publication Date
2026-05-05

AI Technical Summary

Technical Problem

In existing temporal feature knowledge graph logical query methods, fixed logical strategies cannot adapt to the differences in semantic relevance between different branch pairs, resulting in limited logical query reasoning accuracy, especially when dealing with heterogeneous logical structures, where the generalization ability is insufficient.

Method used

By employing a post-order traversal based on a query computation tree, the semantic similarity of the relational feature vectors of the branch pairs is calculated, the weights of the logic system are dynamically adjusted, and a gating mechanism is combined to generate hybrid weights. This constructs a multilayer perceptron (MLP) to automatically learn the complex relationships between branch pairs, thereby achieving differentiated reasoning in the logic system.

Benefits of technology

It significantly improves the accuracy of logical query responses, especially when dealing with data with strong semantic heterogeneity, and enhances the model's generalization ability and inference accuracy.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention discloses a temporal feature knowledge graph logic query answering method, medium and equipment, and the method comprises the steps: taking fixed entity embedding as input according to the post-order traversal of a query calculation tree, carrying out query single-hop projection operation when a projection operator is traversed, and converting entity embedding or query embedding into next-hop query embedding; when negative operators are traversed, complementation is carried out, and when intersection or union operators are traversed, query embedded input of the shunt pairs and relation feature vectors of the shunt pairs are extracted; mapping the semantic similarity of the relation feature vectors into initial weights of the three logic systems; splicing query embedding and relation feature vectors, inputting the spliced query embedding and relation feature vectors into MLP, and calculating dynamic weights of the three logic systems; generating a mixed weight of the three logic systems by utilizing a gating mechanism to obtain query embedding; and calculating a query score according to the query embedding and each entity embedding to obtain a query answer set. The problem that the query precision is restricted by a fixed logic strategy is solved.
Need to check novelty before this filing date? Find Prior Art