Application of Convolution Neural Networks and Randomforest for Software Test

The system development life cycle includes a phase called software testing. Lots of tests are done to test the software. Here, we will focus on how to use Oracle Testing to test system output using black box testing methodology. We will test credit card software as an example, and the results will indicate whether or not a financial institution can authorize credit cards to its customers. This paper explains how to check the output of the credit card application using Convolutional Neural Networks and the Random Forest Algorithm, which will represent the Oracle test, according to the black box testing methodology, using machine learning algorithms, which can be adopted in the next versions of the credit card software to verify the validity of the results. The aim of this paper is to give results about whether a financial institution can approve credit cards for its customers. The decision to approve the card is made by the financial companies based on consideration of various reasons related to individuals which vary from creditworthiness, loans, repayment history and income criteria.