location_city Virtual Platform schedule Sep 12th 12:30 - 01:15 PM IST place Online Meeting 2 people 43 Interested

The adoption of Artificial Intelligence is getting more traction, it is in need to enhance QA capabilities to cope up with these skills. Machine Learning is used extensively in retail applications for solving complex problems, one of them is solving the search relevancy. Showing the appropriate results for the user is important for the conversion rate to go high. As Machine Learning poses different challenges such as a Test Oracle, Fairness, Correctness and Robustness to do QA, We may need to follow different approaches and testing techniques to do the QA for Machine Learning models.


Different Machine learning types such as Supervised and Unsupervised Models have different characteristics and are used for different types of problems. Though these solves different complex problems, Machine learning Models also a unit of software code that needs to be verified as a normal software system. When a Machine learning model is seen as a whole system, it may look complex and unsolvable. We can group them into small modules and verify for quality. Black box and White box testing techniques can be applied to verify the functionality. Data, Feature Engineering and Algorithms are the major part of the Machine Learning model. We will see how we applied different techniques to validate these.


This talk is focused on viewing the Machine Learning software as a whole and performing the Quality Analysis for it. We look at how different is testing a machine learning model from typical software testing. We will discuss the challenges that came across, the Process involved in building an ML model. We take an example of Search Relevance for an explanation. We will dive into the areas where quality is assessed. The significant factors considered here are measuring Accuracy and Efficiency. We will look into the different black box testing techniques for different Algorithms. Let us also see how traditional testing is different from testing machine learning applications. I will go through different black-box testing techniques with examples following a live demo.


Outline/Structure of the Talk

  • Introduction to Machine Learning - 3 mins
    • Machine Learning
    • Problem It Solves
    • Types of ML Algorithms
  • Quality Aspects for ML Application - 5 mins
  • How a Machine Learning Model is Built - 4 mins
  • Search Relevancy Testing - 5 mins
    • What is Search Relevancy?
    • Why did we choose Machine Learning for Search Relevancy Tuning?
    • How to assess the quality of an ML Model.
    • Different steps where QA is applied.
  • Challenges in Testing Machine Learning Model - 10 mins
    • Data/Features/Algorithm
  • Testing Techniques - 13 mins
    • Model Performance
    • Metamorphic testing
    • Fuzzing for ML Testing
  • Comparison between Traditional Testing and ML Testing - 5 mins

Learning Outcome

  • Knowledge of ML and different terminologies
  • Information on how to solve Non-testable problem
  • How to do QA for ML Models
  • Difference between Traditional Software testing and ML Testing

Target Audience

QA, SDET, Developers, Test Lead, Architects, Project Managers


schedule Submitted 2 years ago

Public Feedback

    • Virender Singh

      Virender Singh - Rise of Shadow DOM - Lets solve it through WebDriver

      45 Mins

      Shadow DOM is kind of web standard that developers use to encapsulate their custom HTML code and style components so that no other global style components can override their code. It ensures that a component will work in any environment, even if other CSS or JavaScript is run on the page i.e. Shadow DOM hides away the complexity of the components from the rest of the page because it renders separately from main DOM.

      Selenium legacy selectors are not able to interact these custom elements under the shadow DOM so we need java script to interact with shadow DOM elements and we also require JavascriptExecutor interface for their execution.

      We will also talk about the WebdriverIO tool v5.5.0, which implicitly support Shadow DOM elements.

    • 45 Mins

      Puppeteer is a Node js library (developed by Google Chrome team) to control Chrome and Firefox and is getting lot of traction recently because of its amazing capabilities. It has already become so popular that it has got 50K+ stars on Github against Selenium's 15K+ stars.

      In the last Google I/O event, this tool's capability has been showcased and it has been perceived as the next generation Web Test Automation Tool.

      Is Puppeteer better than Selenium? The answer is 'No', and I will cover 'why' in detail during this talk. I will show some live examples to demonstrate that Selenium can also do all those advance things which Puppeteer promises to do.

    • Sanjay Kumar
      Sanjay Kumar
      ChroPath Creator
      schedule 2 years ago
      Sold Out!
      45 Mins
      Case Study
      • Want to save 70-80% manual effort of automation script writing!!
      • Wasting time in verifying xpath one by one?
      • Want to complete automation script without wasting much time?
      • Are you still wasting time in writing english manual test cases?
    • Pallavi Sharma

      Pallavi Sharma - Learn Selenium with JavaScript

      Pallavi Sharma
      Pallavi Sharma
      5 Elements Learning
      schedule 2 years ago
      Sold Out!
      480 Mins

      Selenium is a popular open source test automation tool, so popular it has its own conference event which run for 3 whole days across continents and is attended by hundreds of enthusiastic testers, and people wanting to know more about testing and automation. In my little experience of coaching for last 7 years, i feel what english language has done for common man[those from non english speaking countries] in earning bread and butter and giving them an elite status, selenium has done for manual testers.

      Anyone and everyone wants to learn Selenium. The magic of selenium or the flexibility is that it can be learnt with your choice of programming languages. I have taught people Selenium with Java, C#, Python and Ruby largely in past these years. I wish to delve into the world of Javascript now and see what this has to offer to us. So this workshop is for those open minded souls out there who wish to pick up another language and learn, unlearn, relearn selenium with it.

    • Srinivasu gangam

      Srinivasu gangam - Zero Touch Automation using NLP (Natural language processing) & AI

      Srinivasu gangam
      Srinivasu gangam
      Delivery Manager
      UST Global
      schedule 2 years ago
      Sold Out!
      45 Mins

      Problem Statement:

      As part of SDLC process:

      1. Is your product quality impacted due to a smaller number of QA resources available in the team?
      2. Are you waiting for QA resources to certify your code every time when you deploy? Is this impacting your product lead time (Speed to Market)?
      3. Is your Product delivery timelines are impacted due to last minute defects identified?
      4. Do you have your QA resources only in one location, but you want to “follow-the-sun” approach for Software delivery across multiple locations?
      5. Do you have manual testers who are not skilled in programming, but you want them to execute automated test scripts w/o any training efforts and automation setup?
      6. Would you like your team more agile and cross functional with Delivery?
      7. Would you like to increase your QA team’s productivity while they invest more time in script development rather than script execution?

      If answer is ‘Yes’ for above questions, "Zero touch automation" is the solution for above challenges that we have been facing part of SDLC.

      Solution: Zero touch automation with cutting-edge technologies

      In this session, I will cover how we solved this problem using innovative solutions, Cutting-edge technologies like NLP (Natural language processing), AI & Cloud solutions.

      You will learn how AI, NLP integrated with core automation components to achieve Zero touch automation.

      This solution is not just revolutionary, it is paradigm shift in test automation to get results to your email with detailed analysis of failure categorization with recommended actions to users.

      I will also cover how E2E automation will be driven with decisions taken by machines based on what user is looking for . There is no manual intervention in this process. NLP and AI play key role to help machines to take decisions.

      We will also cover how we empowered developer/release manager/any team member/Manager to trigger the scripts from their cell phone and get the detailed execution report without having any automation software installed in their computer or Phone.

      We will be demonstrating how the request will be initiated from User, understand the need from user using NLP & AI , Fetching the code from bitbucket to select appropriate automation scripts , running them on Selenoid/docker server , storing results to MongoDB , receiving email with test results and Failure analysis.

      What is the value of zero touch automation?

      1. Enable speed to market: Now that Developers does not need to wait for QA resource, Changes can be certified quickly and ready to push to production. Lead time will be significantly reduced.
      2. Increase quality: Now that test automation is easy and it can run multiple times in each environment, most of the defects will be uncovered and addressed before code goes to production.
      3. Ease of test execution: Test execution will be very easy, no automation or framework setup required from user side. Test execution can be done 24*7.
      4. Productivity: Increase QA team’s Productivity to focus more on script development rather than focusing on script execution and failure analysis .
    • Vishal Srivastava

      Vishal Srivastava / Khyati Sehgal - Setup Grid Infra with AWS Nodes & Autoscaling of AWS Nodes

      45 Mins

      When it comes to creating a project we always search for ideas that are open-source, easily scalable, ready to use.

      We as humans always have a tendency to grow and expand and when it comes to customizable infrastructure it's a big Yes.

      In this talk, we will be focussing on how we can save the cost of actual servers with easily scalable and manageable open-source software using an open-source tool - selenium grid. How we set up the nodes and hub altogether to make cross-browser, multi-platform execution takes place.

    • Krishnan Mahadevan

      Krishnan Mahadevan - My experiments with Grid

      45 Mins

      Everyone starts off with a simple grid setup which involves a hub and one or more nodes.

      This traditional setup is a good start but the moment one starts to get serious with the selenium grid and decide to house their own selenium grid for their local executions, that is when issues start.

      My experiences with the Selenium grid in the past couple of years has led me to get introduced some of the most prevalent problems with maintaining an in-house selenium grid.

      • Nodes get unhooked randomly due to network glitches.
      • Nodes introduce false failures due to memory leaks.
      • Selenium Grid running out of capacity.
      • Nodes require OS upgrades/patches etc.
      • Needing to deal with auto upgrades by browsers (especially chrome and firefox)

      Some of these issues I managed to fix by building a "Self Healing" Grid wherein the nodes automatically get restarted after they have serviced "n" tests. But that still didn’t solve many of these other problems.

      That was when I felt, what if there was an on-demand selenium grid.

      What if the Grid could do the following ?

      • The Grid auto scales itself in terms of the nodes based on the current load.
      • The Grid does not require a lot of infrastructure to support it.
      • The Grid can plug itself into some of the cloud providers or leverage a solution such as Docker so that the nodes can be spun and shutdown at will.

      That was how the idea of "Just Ask" an on-demand grid was born.

      Just-Ask is an on-demand grid. It has no nodes attached to it.

      It’s designed to spin off nodes on demand, run test against the newly spun off node and after test runs to completion, clean-up the node as well. The node can be backed by anything. It could be Docker (or) it could be a VM running on any of the popular clouds.

      The session aspires to walk the audience through with my experiments with the selenium grid, my learnings on the selenium grid internals and how I used all of that knowledge to build my own On Demand Selenium Grid. What better avenue to share these learnings than a Selenium Conference.

      The session will introduce the audience to the grid internals and their concepts such as

      • What is a Selenium Remote Proxy ? What is it used for? What can you do with it?
      • What is a Hub (or) Node level Servlet ? When would you need one ?
      • All of this followed by a quick demo on "Just Ask", the on-demand grid that I have built and open sourced here: https://github.com/rationaleEmotions/just-ask

    • Argneshu Gupta

      Argneshu Gupta - Get rid off Xpath in iOS Native App Automation

      Argneshu Gupta
      Argneshu Gupta
      QA Lead
      schedule 2 years ago
      Sold Out!
      45 Mins

      It is always challenging for us while dealing with Xpath and Css in iOS Mobile Automation.Even we can't think to do iOS native app mobile automation without them .

      So problems we need to face are

      • High Maintenance
      • High Execution Time
      • Complex installation
      • Additional Web server

      Do we have some solutions for this problem - The answer is Yes !!! -> KIF(Keep It Functional)