Knowledge graph construction and improvement system and method based on natural language

Abstract

The invention discloses a knowledge graph construction and improvement system and method based on natural language. The system comprises an original database, a data processing module, and a knowledge graph generation module. The original database is in communication connection with a mathematical processing module, and the data processing module is in communication connection with the knowledge graph generation module. The inference module is used for receiving inference rules, knowledge entities and relationship attributes sent by an original database and a data processing module, generating an inference knowledge graph according to the inference rules, the knowledge entities and the relationship attributes, and sending the inference knowledge graph to the knowledge graph generation module; the knowledge graph generation module receives the reasoning knowledge graph sent by the reasoning module, and fuses the reasoning knowledge graph with the basic knowledge graph to obtain a fused knowledge graph, so that the timeliness of the knowledge graph is ensured, and massive data processing in the knowledge graph updating process is avoided.

Description

technical field [0001] The invention relates to the technical field of knowledge map construction, in particular to a natural language-based knowledge map construction and improvement method. Background technique [0002] At present, the storage in the knowledge map is limited and static, including entity-attribute (or relationship)-entity knowledge triplets, and some knowledge associations are inevitably missing. In addition, the current knowledge graph storage technology ignores the timeliness of knowledge. However, in the application process of the knowledge graph, it needs to be applied to the derived questions that contain the facts described in the knowledge graph, which makes the existing knowledge triples unable to directly answer user questions. For example: "How old are you this year?" The knowledge base stores someone's birthday, but does not contain the dynamic knowledge "age" related to time. But someone's "age" also increases over time. It can be seen that t...

AI Technical Summary

Problems solved by technology

2. Introduce virtual nodes to express the timeliness of knowledge triples, but the introduction of virtual nodes will lead to changes in the knowledge base structure itself

Therefore, whether it is the knowledge graph after adding the state dimension or the knowledge graph after introducing virtual nodes, the compatibility with other knowledge graphs is relatively poor, and it is difficult to realize the integration and interoperability between knowledge graphs

On the other hand, due to the addition of temporal expressions to ensure the consistency and real-time performance of the knowledge base, the calculation and complexity of RDF (Resource Description Framework, resource description framework) triples are greatly increased.

Neither of the above two solutions can well meet the user's needs for knowledge graphs

Images