Transform-based programming problem post title automatic generation method

A technology for automatic generation and problem generation. It is applied in the creation/generation of source code, software engineering design, natural language data processing, etc. It can solve problems such as problem posts that cannot be effectively helped, and achieve the effect of improving performance and quality.

Pending Publication Date: 2022-03-25
NANTONG UNIVERSITY
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to provide a method for automatically generating the title of a programming question post based on Transformer, so as to solve the problem caused by the low quality of the question post title in the process of creating a question post on a programmer's question-and-answer website (such as Stack Overflow) in the prior art The problem that th

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
  • Transform-based programming problem post title automatic generation method
  • Transform-based programming problem post title automatic generation method
  • Transform-based programming problem post title automatic generation method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0039] In order to clearly illustrate the technical features of the solution, the solution will be described below through specific implementation modes.

[0040] see figure 1 , figure 2 , the present invention is a method for automatically generating the title of a programming question post based on Transformer:

[0041] 1. Collection of corpus

[0042] 1.1, such as figure 2 As shown, the question post in Stack Overflow is made up of title, problem description, code snippet, label, score, accepted answer etc., use Java, C#, Python and JavaScript label in the embodiment of the present invention to collect developer question answering website (For example, Stack Overflow) related question posts, in order to collect high-quality question posts, three heuristic rules are proposed to filter question posts.

[0043] 1.2. According to the heuristic rule 1, the score of the question post is not less than five, and it is filtered;

[0044] 1.3. According to heuri...

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 provides an automatic generation method of a programming problem post title based on Transform, and belongs to the technical field of computer application. According to the technical scheme, the method comprises the following steps of: (1) collecting question posts related to four programming languages, namely Java, C #, Python and JavaScript, on a question and answer website of a developer, and filtering out low-quality question posts on the basis of three heuristic rules; and (2) on the basis of the constructed corpus, performing fine tuning on the constructed T5 model based on Transform, and performing optimization by using an AdaFactor optimizer and an attention mechanism to construct an automatic question post title generation model. The title generation method has the beneficial effects that a simple and accurate title is generated for a question post of a question and answer website (such as Stack Overflow) of a developer, and automation of question post title generation for four programming languages of Java, C #, Python and JavaScript is realized.

Description

technical field [0001] The invention relates to the technical field of computer applications, in particular to a Transformer-based method for automatically generating titles of programming question posts. Background technique [0002] Developers can post their questions on developer question-and-answer sites (such as Stack Overflow, or SO for short), and wait for other community members to give corresponding solutions. These questions and corresponding answers can be reused when developers encounter similar problems. So far, millions of developers have used StackOverflow to find high-quality answers to their programming questions. [0003] Although the number of question posts on Stack Overflow is growing rapidly, there are still a large number of questions that do not receive high-quality answers. These questions may not be specific, understandable, or appealing to the relevant developers to answer. These low-quality questionable posts not only do not get effective help,...

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
IPC IPC(8): G06F16/34G06F16/36G06F40/258G06F8/30
CPCG06F16/345G06F16/367G06F40/258G06F8/315
Inventor 陈翔刘珂曲豫宾杨光林浩王昭缪芸
Owner NANTONG UNIVERSITY
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