Software defect prediction method and device, electronic equipment and storage medium

Abstract

The invention discloses a software defect prediction method and device, electronic equipment and a storage medium. Comprising the following steps: extracting metric element features and defect information in software with known defect information to obtain training data; using a linear model as a basic model, using an improved MOEA / D algorithm for carrying out optimized training on model parameters according to training data, obtaining a Pareto solution set, and obtaining a set of defect prediction models, in which the improved MOEA / D algorithm introduces a zero parameter construction mode to optimize the model parameters, and therefore optimization of model performance is achieved; and on the basis of the optimized model, selecting one prediction model according to an application scene to perform defect prediction on the to-be-tested software to obtain defect information of the next version of to-be-tested software. Through the implementation of the method, a software defect prediction model with good sorting performance, relatively small prediction error and relatively short prediction time can be better constructed, and the application scene of the model is further expanded, and the accuracy of defect prediction is improved.

Description

technical field [0001] The invention relates to the field of software detection, in particular to a software defect prediction method, device, electronic equipment and storage medium. Background technique [0002] With the rise of various emerging computer technologies, the scale of software is getting bigger and bigger, and the quality of software is getting more and more attention. Software defects are an important factor affecting software quality and a potential source of errors and failures in related systems. Therefore, , How to discover and correct software defects in time to improve software quality has become a problem that cannot be ignored in the software development life cycle. [0003] Software defect prediction technology started in the 1970s. It mainly uses machine learning technology to generate software defect prediction models based on recorded historical data and currently known defects and other software measurement data, and predicts the future of softwa...

AI Technical Summary

Problems solved by technology

[0004] The main purpose of the present invention is to provide a software defect prediction method, device, electronic equipment and storage medium to solve the problem that the software defect prediction scheme in the prior art has low defect prediction accuracy and complicated prediction models for new software question

Images