Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

Natural language to query by analyzing method

IP.com Disclosure Number: IPCOM000248855D
Publication Date: 2017-Jan-18
Document File: 2 page(s) / 33K

Publishing Venue

The IP.com Prior Art Database

Abstract

Currently, more and more information selected is required, but not everyone could write Structured Query Language(SQL) directly or it is hard for everyone to understand the database object structure. We would like to introduce a method about natural language to sql transformation.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 57% of the total text.

1

Natural language to query by analyzing method

Currently, more and more information selected is required, but not everyone could write Structured Query Language(SQL) directly or it is hard for everyone to understand the database object structure. We would like to introduce a method about natural language to Structured Query Language(SQL) transformation. It would be executed by the following processes: 1. Mark the tag by updating in the catalog. 2. Detect the relationship between table by table definition and workload query relationship 3. Analyzing the natural language to decide whether we could generated the query 4. Check the authority for the query 5. Return the result if the user has the authority The following is the detail for the process for the disclosure.

2

1. Mark the tag during the objects definition or updated in the catalog. For the catalog table, the comments in the catalog table could be defined to indicate the column meaning. The left part is the column name and the right part is the tag

for the table. ENO: Employ number; MNO: Manager employee number

2. Detect the relationship between table by table definition and workload query relationship As the table defined above, there is a foreign key defined on the two tables, there is relationship for the two tables. We could also get the relationship from queries of the workload.

3. Analyzing the natural language to decide whether we could generated the query The system could get the information by natural language...