They also mentioned that if the target class (faulty/non-faulty) is in the minority, accuracy is a poor measure as for example, a classifier could score 90%% accuracy on a dataset with 10%% faulty data, even if it predicts that all defective modules are defect free.
PC3 is the last dataset that has 1,563 modules; 10%% of the data are defected and it belongs to flight software for earth orbiting satellite [39].
The smallest dataset, CM1, belongs to NASA spacecraft instrument project and it has 498 modules and 10%% of the data are defected.
To provide illustrative examples of the definitions listed below, a machine learning application of software module defect classification is used where a learner is trained to predict whether a software module is defect prone or not.
Some Democrats are defecting.
Its politicians are defecting daily.
Liberals who once supported him are defecting.
Indeed, some are defecting from commercial travel.
Superdelegates are defecting.
