Gray generalized regression neural network-based small sample software reliability prediction method

Abstract

The invention discloses a gray generalized regression neural network-based small sample software reliability prediction method. The method comprises the following steps of: first, respectively emulating and expanding failure time data and test coverage rate data in collected small sample software reliability test data by using an improved Bootstrap method to form expanded reliability data which has the same failure statistical rule as small sample reliability data; then, obtaining a three-dimensional curve of the failure time, the test coverage rate and the unreliability of the expanded reliability data; next, establishing a gray generalized regression neural network; later on, training the gray generalized regression neural network by adopting the expanded reliability data and establishing a small sample software reliability prediction model; and finally, predicting by using the model to obtain software reliability prediction information. The method avoids solving a complex multivariate likelihood equation, and solves the problem that an available prediction model can be obtained only by training a large number of models in artificial neural network modeling in software reliability prediction.

Description

Technical field [0001] The invention relates to a software reliability prediction method in software reliability testing, in particular to a small sample software reliability prediction method based on gray generalized regression neural network, and belongs to the technical field of software reliability prediction. Background technique [0002] With the rapid development of computer technology, people's dependence on computer software is increasing, and the requirements for software reliability are getting higher and higher. Software reliability prediction faces the problem of predicting and modeling software with high reliability requirements and a small number of test data. In addition, if the software reliability prediction is to be carried out in the early stage of the software reliability test, there is also a problem that the number of data is small, and it is difficult to meet the sample size requirements of traditional prediction methods. If according to the traditional ...

AI Technical Summary

Problems solved by technology

[0029] The purpose of the present invention is to solve the problem that the prediction accuracy of the prediction model established by the traditional software reliability prediction method in the case of a small sample is low or even the problem that an available model cannot be established. Based on the expected advantages, the test coverage information is added to the model and the improved Bootstrap method is used to expand the small sample data, so as to achieve the effect of high-precision modeling and prediction of software reliability under small sample conditions. Formed a software reliability prediction method based on gray generalized regression neural network considering test coverage under small sample data

Then use the least squares method to fit the probability distribution function of the determined sample to correct the sample empirical distribution function, so as to solve the problem of Bootstrap method resampling in the case of small samples to generate self-service samples without making artificial assumptions on the overall distribution. Reasonable questions; finally construct failure time-unreliability curves and test coverage-unreliability curves to integrate test coverage into prediction models

Images