Workload perception defect prediction method based on weighted software network

Abstract

The invention belongs to the field of software defect prediction, and particularly discloses a workload perception defect prediction method based on a weighted software network, which designs a correlation strength calculation method among software modules according to two correlation relationships of dependence among the software modules and collaborative developers. Constructing an effective weighted software network structure, autonomously learning feature representation of software modules in a weighted software network graph by adopting strong learning ability of a graph embedding technology, and better reflecting data among the software modules and calling dependency and dependency relationships of collaborative developers; meanwhile, the workload of the inspection code for discovering the defects is considered in the construction of the defect prediction method, the actual requirements of software development are met, and the software defects can be conveniently, quickly and accurately discovered.

Description

technical field [0001] The invention belongs to the field of software defect prediction and relates to a workload-aware defect prediction method based on a weighted software network. Background technique [0002] There are inevitably defects in the software development process, and these defects will lead to serious economic losses. Therefore, how to quickly and accurately find software defects plays a vital role in ensuring the quality of software systems. [0003] Software defect prediction technology is committed to identifying high-risk defective modules, narrowing the scope of developers' review and testing code, and realizing the reasonable allocation of limited resources. The defect prediction method based on metric information is the most commonly used method, and its metric meta information mainly includes artificially designed metric elements and self-learning metric elements based on abstract syntax trees. [0004] However, these features based on abstract synta...

AI Technical Summary

Problems solved by technology

[0012] 1. When building a software network, the impact of the correlation strength between modules on defect identification is not considered;

[0013] 2. Lack of workload perception module, the obtained defect classification results still need a lot of time to review the code, lack of operability

[0014] Software defects seriously affect the quality of software, and even cause serious economic losses

Images