Introduction
Machine Learning is growing every day in all sorts of areas and some examples are for e.g., voice recognition systems used by virtual personal assistants like Amazon Alexa or Google Home. As the field of ML continues to grow there needs to be more and more need for the quality assurance to make sure everything works as expected. Reliability is another key need.
Things to Consider:
Software researchers have started adapting concepts from the software testing domain (e.g., code coverage, mutation testing, or property-based testing) to help ML engineers detect and correct faults in Machine Learning programs.
Checklist:
- Make sure that your testing use a tool that can call most important methods or APIs for appropriate code coverage.
- Some sort of automated log checks for errors and repetitive ones are looked into by doing some digging
- Sentimental analysis of the discussion should be looked upon to see if the user is happy after the interaction
- Periodic surveys are another powerful tool
- Different interface for testing would help for examples Web User Interface, Chat bot interface, Direct NLP techniques etc
Author Bio
Shalu Chawla works in the field of Automation Engineer, QA Analyst and Business Analysis. For more about her please visit her website www.shaluchawla.com . You can also visit me in my other website on wellness by clicking on https://shalu-chawla.org/ if interested in meditation or https://shalu-chawla.net/ if interested in travel topics.
Software QA and Testing 