Laveena shares how she was able to explore the world of data science as a tester when testing a model and how you can apply that if you find yourself in a similar situation. As part of an emerging team, how she was able to contribute value in a new field that she had never tested before.

Having heard from other senior testers that they know of data science teams but no testers testing the models, how do we have enough confidence what is produced is good enough? A model is a statistical black box, how do we test it so we understand its behaviors to test it properly? The main focus should be to help inspire testers to explore data science models.

Join Laveena as she goes through a journey of discovering data science model testing and finding the following takeaways useful, not just for testing a data science model, but for day-to-day testing too.

1. Some background of what a data science model is, and how data plays a role in these models. Understand from vast amounts of data, including:
• structured data
• unstructured data
• metadata
• semi-structured data
2. Understand data pipelines.
3. Importance of pairing – Follow a SDLC process which may require a bit more of exploratory testing and investigation, therefore pairing with data scientist is a good way of working and understanding the model.
4. Pre-testing thoughts: is the model custom made or off the shelf? How as a team are you training your own model to behave? What is your input and what’s your output? Are you experiencing the right behavior? (models do contain some element of randomness so how will you make sure what’s acceptable when testing the results?)
5. As testers we expect input + model that uses predictive analytics = output (example 5+3 = 8), but for data scientists “5+3” does not always equal 8 but 8.1, 8.001, 8.5, etc. In simple words – stochastic. So how will we bring processes and strategies to make sure we capture the right output results and the consumers still benefit from this? In a nutshell, making sure the model’s quality is good and we have the confidence in what we provide to consumers.
6. Test the areas we are certain about the behavior and those areas uncertain about have some bounds around averages – expectations set.
7. Exploratory testing and looking for edge cases, regression testing to see that new features are not breaking baseline results.
8. Understanding what tests to perform: what is an acceptable test for the model? Have we found anomalies? (results too off the threshold?) How do we know what we produced as results is the right result? How accurate are the results from the model? What is an acceptable deviation?
9. Laveena will give away tips that helped her and could help testers who want to explore testing models and making sure the quality of a model is providing the team enough confidence and helping a business.
10. Post testing – Have we got a good understanding of what the model has provided? Are the predictive analytics working as expected? Does the shape of the data look as expected? (testing the outputs will explain if the values are of the right type from the data input stage).

Webinar Takeaways:

• Have a better understanding of what data science is.
• Know how we can test models.
• Know what existing skills we already have that we can apply in a data science team.
• Leave with resources to help our teams’ better structure itself to have confidence in the data produced.
• We’ll look at what did or didn’t work.

Webinar Speaker:

Laveena Ramchandani – Senior Consultant – Testing, Miss
Laveena is an experienced Testing Consultant with a comprehensive understanding of tools available for software testing and analysis. She is an energetic, technical-minded professional seeking a position as a Product Experience Analyst. Her aim is to provide valuable insights that have high technical aptitude, and unyielding commitment to work. Being able to inspire more individuals out there in the world be a great achievement.

Speaker Details:

Joining from London, United Kingdom
LinkedIn: Laveena Ramchandani