Abstract
Much software lacks test oracles, which limits automated testing. Metamorphic testing is one proposed method for automating the testing process for programs without test oracles. Unfortunately, finding appropriate metamorphic relations for use in metamorphic testing remains a labor intensive task, which is generally performed by a domain expert or a programmer. We are investigating novel approaches for automatically predicting metamorphic relations using machine learning techniques. Preliminary results show that the proposed techniques are highly effective in predicting metamorphic relations.