Design of Automated Testing Method for Information Management Software
To improve the accuracy of anomaly identification of information management software,an automated testing method for information management software is designed based on deep neural network.Markov chain is utilized to construct an information management software task profile model to extract the task flow for automated testing of information management software.Deep priority traverses the profile to fuse the information of change conditions,post conditions,and operation probabilities of automated test operation behaviors.On this basis,extract the task flow for automated testing of information management software.Combine the known software task state transfer probability to calculate the limit probability,and take the use cases and parameter information with significant importance as the information of priority testing to complete the network training.Establish the software anomaly automation identification model,locate the software anomaly location according to the anomaly identification results,and complete the information management software automation test.The test results show that the software automation test results of this method are more accurate,and the automation efficiency is higher,which has certain application value.
Deep neural networksInformation managementSoftwareAutomationTestingSelf-encoder
万方数据
维普
