Invention Grant
- Patent Title: Natural language processing engine for translating questions into executable database queries
-
Application No.: US16866034Application Date: 2020-05-04
-
Publication No.: US11573957B2Publication Date: 2023-02-07
- Inventor: Jichuan Zeng , Xi Lin , Chu Hong Hoi
- Applicant: salesforce.com, inc.
- Applicant Address: US CA San Francisco
- Assignee: salesforce.com, inc.
- Current Assignee: salesforce.com, inc.
- Current Assignee Address: US CA San Francisco
- Agency: Haynes and Boone, LLP
- Main IPC: G06F16/242
- IPC: G06F16/242 ; G06F16/2452 ; G06F40/47 ; G06F40/284

Abstract:
A system and method for translating questions into database queries are provided. A text to database query system receives a natural language question and a structure in a database. Question tokens are generated from the question and query tokens are generated from the structure in the database. The question tokens and query tokens are concatenated into a sentence and a sentence token is added to the sentence. A BERT network generates question hidden states for the question tokens, query hidden states for the query tokens, and a classifier hidden state for the sentence token. A translatability predictor network determines if the question is translatable or untranslatable. A decoder converts a translatable question into an executable query. A confusion span predictor network identifies a confusion span in the untranslatable question that causes the question to be untranslatable. An auto-correction module to auto-correct the tokens in the confusion span.
Public/Granted literature
- US20210173829A1 Natural Language Processing Engine For Translating Questions Into Executable Database Queries Public/Granted day:2021-06-10
Information query