Natural Language to SQL Methods, Devices, and Storage Media Applied to the Securities Industry
By utilizing knowledge graph matching and triplet extraction techniques in the securities industry, a natural language to SQL conversion method specific to the securities industry is generated, solving the problem of insufficient accuracy of natural language to SQL conversion methods in the securities industry and achieving higher accuracy and reliability.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- HUAAN SECURITIES CO LTD
- Filing Date
- 2026-03-13
- Publication Date
- 2026-05-26
AI Technical Summary
Existing natural language to SQL conversion methods are difficult to guarantee accuracy in the securities industry, especially when faced with complex and ever-changing natural language and massive securities industry data, making it difficult to generate accurate SQL query statements.
By receiving natural language data from the securities industry, the system uses a pre-defined knowledge graph to generate a target sub-graph, extracts entities and relationships from triples, filters relevant text fragments based on weight information, fills them into a table template to generate tabular formal language, and finally generates accurate SQL statements.
This improves the accuracy of natural language to SQL conversion methods in the securities industry, ensuring that the generated SQL statements are more reliable and accurate, and adapting to the complex data needs of the securities industry.
Smart Images

Figure CN121833757B_ABST