In software development, the software testing phase is an important process, one of which is the determination of the classification of fault output classes. This paper presents the results of the design and implementation of the integration-based model (I-BM) framework as a software testing framework. The integration-based model (I-BM) framework has the ability to systematically classify software fault output classes in the form of datasets. The research method itself is experimental. This integration-based model (I-BM) framework will document the tested software's fault output based on variables such as function, interface, structure, performance, requirement, documentation, positive, and negative. This I-BM Framework can also be used by software companies' testing divisions. In the final stage of the I-BM framework, the accuracy level of the Fault Output is measured by comparing the fault output of Neural Network Algorithm, SVM, and I-BM to the Actual Expected Fault Output. The average accuracy levels are 0.86, 0.81, 0.85, and 0.85.