Software code recommendation method based on JEAN

Abstract

The invention provides a JEAN-based software code recommendation method, which is mainly used for solving the problem of recommending code segments to developers in software development. The method comprises the following steps: S1, constructing a JEAN model; S2, collecting a java code segment data set with description from GitHub, and pre-processing the java code segment data set as a training set for training a JEAN model; S3, constructing a code library, and embedding all code segments in the code library into code vectors by using a code embedding representation module in the JEAN model; S4, performing online code search by a developer , and after query is input, embedding the query into a query vector by a description embedding representation module in the JEAN model; and S5, calculating the similarity between the query vector and all code vectors in the code library by using cosine similarity, and returning a code segment of the vector most relevant to the query vector. According to the method, software developers can be effectively helped to recommend proper code segments according to own requirements, and the method has relatively high accuracy and efficiency.

Description

technical field [0001] The invention belongs to the technical field of information retrieval, and in particular relates to a JEAN-based software code recommendation method. Background technique [0002] In the process of software development, in order to realize certain program functions, developers often have to perform some unfamiliar programming tasks. When developers encounter these programming problems, they tend to search the web for code snippets that can solve these programming problems. However, studies have shown that software developers spend an average of about 21% of their development time on web searches. This is because with the continuous development of Internet-related technologies, the speed of information dissemination has also increased significantly, resulting in an increasingly serious problem of information overload. , It is becoming more and more difficult to screen out valuable information from massive amounts of data. [0003] Developers will sear...

AI Technical Summary

Problems solved by technology

However, studies have shown that software developers spend an average of about 21% of their development time on web searches. This is because with the continuous development of Internet-related technologies, the speed of information dissemination has also increased significantly, resulting in an increasingly serious problem of information overload. , it is becoming more and more difficult to screen out valuable information from massive amounts of data

[0003] Developers will search for the code snippets they need through search engines or online question-and-answer forums. However, there will be many low-quality or unsuitable answers in the code snippets retrieved through search engines or online question-and-answer forums.

Images