Software problem report classifying method based on text chaos degree

Abstract

The invention discloses a software problem report classifying method based on the text chaos degree. The software problem report classifying method comprises the following steps that 1, data acquisition is conducted on a GitHub platform, wherein acquired data includes problem reports submitted by the public and relevant information of the reports; 2, datasets are marked, tabs capable of representing the types of the problem reports are selected through artificial sampling analysis on tabs of the problem reports, and accordingly dataset marking is achieved through the tabs; 3, the datasets are preprocessed, including text information except undesired information, and are converted into digital types for calculation processing through text information digitization; 4, model establishment is performed, wherein a two-layer automation problem report classifier is established by utilizing an existing machine learning method. By adopting the method, the automation level of problem report classification can be improved, labor cost is reduced, and the contribution converging efficiency of the public is improved.

Description

technical field [0001] The invention relates to a method for classifying software problem reports based on the degree of confusion in the paper. With the development of the open source community, a large amount of software problem report data has been accumulated in the problem management tools in the open source community. These reports contain multiple types that are mixed together indistinguishably, such as software defects, feature requirements, and so on. Good problem classification plays a huge role in the management of development tasks, but traditional manual classification methods are expensive, so automated methods are of great help to reduce maintenance costs and improve software development and maintenance efficiency. Background technique [0002] With the promotion of the open source movement and the support of development tools, more and more people participate in the development of open source software and provide their own contributions. When making public ...

AI Technical Summary

Problems solved by technology

In addition, the quality of problem reports has also been affected to a certain extent because the threshold for submitting reports has been lowered

These factors have greatly affected the training process of traditional classifiers, and cannot adapt well to the new features of existing problem reports.

Images