Software defect prediction method based on generative adversarial network and ensemble learning

Abstract

The invention discloses a software defect prediction method based on a generative adversarial network and ensemble learning. The method comprises the following steps: 1, carrying out the preprocessing of a software defect data set, dividing the data set into a training set and a test set, and calculating a resampling rate; 2, constructing a generative adversarial network model; 3, inputting the training set into the generative adversarial network for training to obtain a trained generative adversarial network; 4, using the trained generative adversarial network to generate new few-sample defect data according to the resampling rate, and obtaining a resampled training set; and 5, constructing a software defect strong classifier by using an AdaBoost method, and inputting the test set into the trained software defect strong classifier to obtain a software defect prediction result. According to the software defect prediction method, the problem of software defect data imbalance is solved, and the accuracy, the correct rate, the recall rate and the F-measure performance of the software defect prediction method are improved.

Description

technical field [0001] The invention belongs to the technical field of software defect prediction, and in particular relates to a software defect prediction method based on generative confrontation network and integrated learning. Background technique [0002] With the popularization of informatization, there may be software defects in all kinds of software that are in use or in the process of development. Being able to find and locate software defects in a timely manner plays a major role in the normal operation of the software system and the improvement of software functions. . Software Defect Prediction (SDP) aims to find the module most likely to contain defects from a software project, where the module can be a function, a loop body, or a class, etc. Generally, software defect prediction methods are divided into the following three steps. First, according to the vulnerability reports of each module in the historical software, the software modules are marked with defec...

AI Technical Summary

Problems solved by technology

Among them, the SMOTE and ADASYN methods adopt the form of artificially synthesizing few-sample data to reduce the problem of data class imbalance. The rules for generating data are artificially stipulated, and it is difficult to find suitable rules in real software defect data sets.

However, the ROS and RUS methods directly reduce the amount of data in the training set of the software defect prediction model, and the actual application effect is poor.

Images