Knowledge map extension model, structured knowledge storage method and device

Abstract

The invention relates to the field of artificial intelligence, in particular to a knowledge map expansion model, a structured knowledge storage method and device, aiming at solving the problem that the quantized data cannot be stored in the prior art. The knowledge map expansion model of the invention comprises entities and the relationships. The connection between the entities is a directed connection, and the directed connection edges contain the relationships. Each relationship includes one or more optional attributes; each entity, relationship, and attribute has a unique identifier. Each relationship includes one or more optional attributes; each entity, each relationship, and each attribute has a unique identifier; each attribute includes a data type and value, as well as optional dimensions. The storage method of the invention is based on the knowledge map expansion model, stores the structured knowledge as the data table structure of the relational database, stores the quantizeddata into the attribute, facilitates the logical reasoning and the quantified reasoning, and improves the query efficiency.

Description

technical field [0001] The invention relates to the field of artificial intelligence, in particular to a knowledge graph extension model, a structured knowledge storage method and equipment. Background technique [0002] With the development of computer technology and artificial intelligence technology, how to let computers automatically store and understand knowledge, and then perform automatic logical reasoning on knowledge has become a hot research field in recent years. [0003] Ontology is a modeling method for knowledge, which defines each concept as an entity, and defines the connection between entities through a series of standard grammars. The knowledge graph (Knowledge Graph) theory models knowledge in the form of a directed graph, and regards each entity as a node of the graph, and the relationship between entities is the edge connecting the entity nodes. [0004] figure 1 It is a schematic diagram of the entity-relationship model of the existing knowledge graph...

AI Technical Summary

Problems solved by technology

[0007] It can be seen from Table 1 that the existing methods for modeling knowledge graphs can only reflect the qualitative logical relationship between entities for the relationship between knowledge concept entities, but cannot reflect the quantitative relationship between entities, and only for logical reasoning. Able to perform qualitative reasoning, unable to perform quantitative reasoning and more precise quantitative calculations

For example, the existing knowledge graph modeling method can model the knowledge "carrots contain vitamin C", but cannot effectively model the knowledge with quantitative parameters, such as "13 mg of vitamin C per 100 grams of carrots". It is also impossible to make automatic inferences on propositions such as "which foods contain more than 10 mg of vitamin C per 100 g and have calories within 100 kcal"

Images