Recently I wrote about using AI as an assistant in Context-Driven testing. We will explore more in this blog about this topic, as we delve into the topic of “Software Testing: Machine Learning For Prioritizing Tests”, and look into a bit more detail of how to use Machine Learning for the same.
Areas to look for
Your machine learning model should ideally be looking into the following areas for prioritizing tests
- Requirements for the current release
- New features
- Fresh updates to existing features
- Historically known areas for critical issues
- Known defects – resolved and existing with priorities assigned based on supervision by software testers
- User behavior data
- Existing test scenarios
Prioritization of test areas
Based on the areas listed above, the model would prioritize areas of test. Note that this is not prioritization of ‘test cases’, nor it is a list of retests to be run after a change, or a regression test for fixes! It would just be an overall prioritization of test areas to consider so that the testers can focus on those areas for thorough testing. It is important that the testers decide the detailed test areas to be considered, and also supervise whether the areas suggested by the model are realistic and correct.
Conclusion
Hope this blog gave you a glimpse on the topic of “Software Testing: Machine Learning For Prioritizing Tests”. For implementation details about how to go about the same for your organisation, please feel free to get in touch with me. Glad to help.