Tracking down high coverage configuration using clustering and fault detection

Mostly all software systems are highly configured, it has many benefits but there is a difficulty of software testing because there will be unique errors could be hidden in any of the configurations and undergoing testing for each of the configurations will lead to expensive testing and it is also impractical. The dependable systems will have some mechanism for fault tolerance in software testing. If the rate of the fault detection is calculated then the coverage of the configuration can be easily generated. First load the application for which it is going to be tested by using our test case prioritization approach and loading the dataset for the test case for the given application. After this process, need to assign the individual ids for all the test cases in the test case dataset. Also it is able to add the test cases in the dynamic nature. Then to compute the test case prioritization, first built the dependency structure for the test cases. Through the approach get the height and weight matrix for the test cases after this computation the test cases. The cosine similarity values between the test cases. In the similarity values it will show how it is highly related with the other test cases. Thus the clustering approach is introduced for grouping the test cases. These test cases are analyzed for measuring their relevancy and relationship between the test cases using their constrains and the clustering of the test cases is done for the better result in the rate of fault detect. With the Average percentage fault detection the graph is drawn and it shown the high coverage configurations.