AISQL parser in decision-making distributed database system and implementation method thereof

A distributed, database technology, applied in the field of SQL parser, can solve problems such as demanding requirements, project implementation cost, high risk and cycle, lack of support for grammar parsing, etc.

Active Publication Date: 2019-01-18
北京傲韦科技有限公司
View PDF14 Cites 8 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0002] Existing SQL parsers support standard SQL core syntax and have their own specific extended function syntax, but lack support for syntax parsing of AI model management and application-related operations
This makes users have to resort to AI development toolkits to make up for this problem, but AI toolkits are provided to users in the form of programming APIs, which requires users to have some kind of general programming language (such as C / C++ / Python / Java) development Ability, the learning curve of these languages ​​is significantly higher than that of the declarative query language SQL, which is too demanding for traditional data analysts who use SQL as the main analysis method
In addition, there are problems of compilation, packaging, testing, and deployment in the secondary development of AI tools, and the project implementation cost, risk, and cycle are much higher than the traditional data analysis process

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • AISQL parser in decision-making distributed database system and implementation method thereof
  • AISQL parser in decision-making distributed database system and implementation method thereof
  • AISQL parser in decision-making distributed database system and implementation method thereof

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0025] Embodiment 1 of the present invention provides an AI SQL parser in a decision-type distributed database system, such as figure 1 As shown, the SQL parser supporting AI SQL includes:

[0026] The grammar definition module 10 is used to define the grammar for implementing AI SQL;

[0027] The AI ​​SQL parsing module 20 is used to parse the input AI SQL statement to obtain a parsed syntax tree;

[0028] The logical execution plan generation module 30 is used to create a new logical execution plan based on the parsed syntax tree.

[0029] In the present invention, new grammatical definitions are used to implement AI SQL and AI SQL, which is an extended SQL of the SQL2011 standard, reduces the threshold for AI use and the demand for AI programming developers, saves the cost of AI projects, and makes full use of the entire While clustering the overall resources (GPU, CPU, memory, etc.), complex data analysis tasks can be optimized and executed as a whole. Compared with the existing s...

Embodiment 2

[0031] A parser for AI SQL in a decision-type distributed database system, such as figure 2 As shown, the difference from Embodiment 1 is that the grammar definition module 10 includes:

[0032] The grammar definition unit 101 created by the AI ​​model is used to define the grammar created by the AI ​​model,

[0033] ::=CREATE[ ]MODEL

[0034] AS

[0035] ::= TEMPORARY

[0036] ::=GLOBAL|LOCAL

[0037] ::=

[0038] ::= As defined in the SQL 2011 specification

[0039] ::=[ ]

[0040] ::= As defined in the SQL 2011 specification

[0041] ::=.

[0042] ::= As defined in the SQL 2011 specification

[0043] ::= As defined in the SQL 2011 specification;

[0044] The grammar definition unit 102 of the AI ​​model update is used to define the grammar of the AI ​​model update,

[0045] ::=UPDATE[ ]MODEL

[0046] AS ;

[0047] The grammar definition unit 103 of AI model evaluation is used to define the grammar of AI model evaluation,

[0048] ::=SELECT FROM [0049] ::= As defined ...

Embodiment 3

[0062] ::=

[0063] ::= |

[0064] ::= ARRAY

[0065] ::= MULTISET

[0066] among them The middle is the domain data type built into the feature knowledge base by the system, such as age, address, email address, gender, mobile phone number, ID number, hundred-point score, five-point score, altitude, etc. Is the user-extended domain data type;

[0067] The grammar definition unit 105 of the AI ​​model application is used to define the grammar of the AI ​​model application. The application of the AI ​​model is a prediction function that generates a prediction result based on the data set input by the user and the selected built model. The prediction result includes : Classification results, trends, association relationship mining, recommendation results, etc. These results are also a set. In special cases, this set may only have one row of data.

[0068] :=SELECT FROM [0069] ::=[ ] ; [0070] AI SQL other grammar definition unit 106, used to define other grammars of AL SQL...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention belongs to an SQL parser supporting AI SQL, in particular to an AI SQL parser in a decision-type distributed database system and an implementation method thereof. The present invention provides a new AI SQL parser in decision-type distributed database system and an implementation method thereof, The parser and method adopt AI SQL as the extension of SQL 2011 specification, Compared with the secondary development of AI analysis program, Because of the inherent declarative nature of the SQL language, AI SQL much is easier to master and use than other AI programmers such as Python / Java / C / C + +, so that the AI SQL can be mastered by data analysts who are generally proficient in SQL, without the need for additional AI programmers proficient in Python / Java / C / C + +. And AI SQL usesdecision-making data as the execution platform, compared with the re-development process of AI analysis program, it reduces the compilation, packaging, deployment links, testing work is simpler, thusreducing the AI analysis project personnel costs and implementation cycle.

Description

Technical field [0001] The invention belongs to a SQL parser supporting AI SQL, and particularly relates to an AI SQL parser in a decision-type distributed database system and an implementation method thereof. Background technique [0002] The existing SQL parser supports the standard SQL core grammar and has its own specific extended function grammar, but lacks support for the grammar analysis of AI model management and application-related operations. This makes users have to resort to AI development toolkits to make up for this problem, but AI toolkits are provided to users in programming API mode, which requires users to have a certain general programming language (such as C / C++ / Python / Java) development Ability, the learning curve of these languages ​​is significantly higher than that of the declarative query language SQL, which is too demanding for traditional data analysts who use SQL as the main analysis method. In addition, the secondary development of AI tools has proble...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G06F16/2453G06F16/2455
Inventor 张德辉
Owner 北京傲韦科技有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap