Neural Network-based Test Case Prioritization in Continuous Integration NEUTRON-Dataset

The dataset is constructed starting from the three industrial dataset, however, it is not organized around the CI cycles but around a test case.The three industrial datasets are: two from ABB Robotics Norway (https://new.abb.com/products/robotics, Paint Control and IOF/ROL, for testing complex industrial robots) and Google Shared Dataset of Test Suite Results (GSDTSR, https://bitbucket.org/HelgeS/atcs-data/src. All three datasets contain information about historical test case executions, along with the verdicts (pass, fail), with CI cycles over 300. The two ABB datasets are split into daily intervals, whereas GSDTSR is split into hourly intervals as it originally provides log data of 16 days. However, the average test suite size per CI cycle in GSDTSR exceeds that in the ABB datasets. We extracted and reorganized the information provided in the three datasets in order to structure the information around a test case, thus characterizing each test case with execution cycles, fails versus passes, etc. We have obtained similar information as in the initial datasets, except the following: 352 cycles for Paint Control, 89 test cases for Paint Control and 1941 test cases for IOF/ROL.