Evaluating to software quality and testing process based on improved Bass model

User acceptance test(UAT) is an important part of ensuring G Bank Software Quality. Insufficient testing will trigger production problems, while excessive testing causes waste of resources. Quantitative measurement and forecast of test process and test quality is necessary in order to ensure effective use of test resources. Based on the development mode of G bank software, the number of test cases, the number of function points, the number of demand changes, the number of related systems and the development workload were selected as the release characteristic indexes according to the characteristics of the iteration of the short cycle version of bank software. The weight of the index was determined by the entropy method and the defect potential was calculated. At the same time, the estimation method of model coefficient to solve the short-period characteristics is proposed. Therefore, the Bass model is improved by the feature gap between the versions based on the characteristics of G Bank software development mode and the characteristics of the short-period version of the bank software. The improved Bass Model supplies a solution to quantitative evaluation about the current version of the software testing process and test quality based on improved Bass model by using version feature gap and the date of software defect found in the historical testing process, which provides theoretical basis for the reasonable arrangements of software test activities and the criteria for stopping testing. The empirical results indicate that the prediction accuracy of the model is reliable in the internal management software defect quantity trend of waterfall development mode and monitoring the testing process, which greatly improves the rationality of the arrangement of testing resources and testing activities while ensuring the testing quality.