A Software Reliability Evaluation Method Based on Rough Set

Quantitative software reliability evaluation method depends on the failure data of software. And most evaluation work is started only after obtaining the failure data in the later period of software development, which leads to a problem that people could not obtain the information of software reliability level in the early stage of software development, which could be used for guiding the work of software reliability design, analysis and testing. Therefore, the qualitative software reliability evaluation methods based on the reliability corresponding factors in each stage of the software development are studied recently. In this paper, a qualitative software reliability evaluation method based on rough set is proposed, which takes use of the data mining technology based on rough set to analyze the data related to the software reliability, and finds the factors and the correlation rules which have an important influence on the software reliability. Then it could be used to evaluate the software reliability.