SlideShare uma empresa Scribd logo
1 de 31
#ATAGTR2019
Assuring Quality for AI based applications
Vinod Sundararaju Antony, Senthilkumar
Thirumalaisamy, Santhosh Kumar Vasudevan
14th 15th Dec 2019
#ATAGTR2019
As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial
channels(Provided due credit is given to me/us)
Abstract
Key challenges include getting the right
data sets for testing, predicting and test
designing the expected outcome and
behaviour, identifying apt algorithms
AI IS NEARER TO
MAINSTREAM
Intelligent software and
applications are omnipresent and
are changing the way we engage
Artificial Intelligence (AI) technology is drastically
finding its way into conventional software
development
A Robust test strategy which increases the level of confidence in AI Apps is the need of the hour
2
#ATAGTR2019
As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial
channels(Provided due credit is given to me/us)
AI Is Here To Stay
Global AI Business Value to Reach $1.2 Trillion in 2018 and $3.9 trillion in 2022
AI has already picked up pace across various industry sectors
BANKING & FINANCE HEALTHCARE
MARKETING
AUTOMOBILE
Travel concierge
Personalized platform for
Airlines
TRAVEL & CUSTOMER RELATION
SMART HOME
INFORMATION, MEDIA & ENTERTAINMENT
VIRTUAL ASSISTANT
RETAIL
3
#ATAGTR2019
As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial
channels(Provided due credit is given to me/us)
AI Is Nearer To Mainstream
Democratized AI – AI will become more widely available due to cloud computing, open source and the “maker”
community (Deep Neural Nets is just 2-5 years from mainstream adoption)
4
#ATAGTR2019
As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial
channels(Provided due credit is given to me/us)
But There Is Still A Long Way To Go
Google Apologizes for Photo App’s
Racist Blunder
Tesla Car that crashed was running on
Autopilot
Microsoft silences its new AI bot Tay
after Twitter users teach it racism
5
#ATAGTR2019
As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial
channels(Provided due credit is given to me/us)
The spectrum of AI & the challenges in assuring Quality
MACHINE LEARNING
Deep Learning
Predictive Analytics
NATURAL LANGUAGE PROCESSING (NLP)
Translation
Classification & Clustering
Information Extraction
SPEECH Speech to Text
Text to Speech
Expert Systems
Robotics
Vision
Image Recognition
Machine Vision
Artificial Intelligence
• Non-deterministic and probabilistic - No defined input and output
• Ever-changing Behaviour – AI systems are always learning
• Non-Linear inputs – e.g. Voice, conversational/free-flowing text
6
#ATAGTR2019
As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial
channels(Provided due credit is given to me/us)
7
 Detecting credit card fraud
 Determining whether a
customer is likely to switch to
a competitor
 Deciding when to do
preventive maintenance on a
factory robot
What is Machine Learning?
Machine Learning focuses on data-driven predictions as opposed to following strictly
static program instructions
Uses the patterns to predict the futureFind patterns in data
#ATAGTR2019
As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial
channels(Provided due credit is given to me/us)
8
Machine Learning – How does it work ?
Machine
Learning
Algorithm
Model
Application
Data
Contains
Patterns
Finds
Patterns
Recognizes
Patterns
Supplies new
data to see if it
matches
known patterns
Why is it gaining traction ?
Machine Learning requires Lots of data, Lots of compute power, Effective Machine Learning Algorithms.
All of these are more available than ever with the evolution of Big Data, Cloud etc.
#ATAGTR2019
As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial
channels(Provided due credit is given to me/us)
9
What are the testing challenges?
ML applications are
intended to learn
properties of data sets,
expected output is NOT
already known to users
Periodic learning
leading to changing
behaviour over period
of time
Prediction of all
scenarios is a time
consuming process
Dependency on
humungous amount of
data?
#ATAGTR2019
As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial
channels(Provided due credit is given to me/us)
10
Testing a Machine Learning Model
Features make the most important part of a machine learning application or model. Testing features are key set of Test
tasks which needs to be performed for ensuring the high performance of machine learning.
#ATAGTR2019
As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial
channels(Provided due credit is given to me/us)
11
1. Testing Features of Machine Learning models
Feature Thresholds
Feature Relevance
Feature Relationship
Feature Suitability
Feature Compliance
Features Unit Testing
Feature Static Review
Test if the feature relationship with outcome variable in terms of correlation coefficients.
Test whether value of features lies between the threshold values
Age of Human (Y) - Threshold - 0 (X1) to 100 (X2). Test if Y lies between X1 and X2 (or) Y > X1
Test whether the feature importance changed with respect to previous test run.
Test/review the static code analysis outcome of code generating features
Test/review the code coverage of the code generating features
Test/review if the generated feature violates the data compliance related issues
Test the feature unsuitability by testing memory usage, inference latency and more.
#ATAGTR2019
As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial
channels(Provided due credit is given to me/us)
12
2. Testing for metamorphic relations
Test Case 1
X=X1
Machine
Learning
System
Under Test
Output 1
Y1
Test Case 2
X=G(X1)
Output 2
Y2
Input Output
Metamorphic Relations
Input
Relation
Output
Relation
AN AI BASED APPLICATION TO PREDICT THE RISK OF DIABETES
USE CASE
Age BMI Predict Diabetes (Y/ N)
30 30 Y
40 32 N
Indicates failure in
Metamorphic relation
• Validation of relations between the outputs of multiple inputs can help detecting defects in ML algorithm
• Follow-up test cases and validation of the test results and could be fully automated.
#ATAGTR2019
As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial
channels(Provided due credit is given to me/us)
13
2. Testing For Metamorphic Relations – Orientation & Color Testing
ORIENTATION
COLOR
• RGB Channels of an image
constitute pixel values for ‘red’,
‘green’ and ‘blue’ colors
• Permutation of RGB input
channels helps in identification
of metamorphic relations and
detect failures
• Core property of CNN is NOT
violated by RGB permutation
Original, RGB BGR
CNN
SUT
BOATTest case #1
Convolutional
Neural Network
Test case #1 CNN
SUT
BOAT
Convolutional
Neural Network
Test Case #2 –
MR1
Test Case #3 –
MR2
90o 180o
• Facebook uses CNN for
automatic tagging
algorithms,
• Amazon  for generating
product recommendations
and
• Google  for search through
among users’ photos
Test Case #2 –
MR1
Test Case #3 –
MR2
#ATAGTR2019
As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial
channels(Provided due credit is given to me/us)
14
Use Case & Demo – Image Based Search in Social Media / Retail
34% 85%
Consumers will spend more money online
when AI is deployed effectively
Customer interactions in retail will be
managed by artificial intelligence, by 2020
Pinterest LensHow it should ideally work ? A Real life scenario
#ATAGTR2019
As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial
channels(Provided due credit is given to me/us)
15
Use Case & Demo
pickle
Serialize the data
Trained Model
Train dataset
Test Dataset for
prediction
Shoe Images 60K
predict
ML Application
CNN - ResNet
TESTING FOR MACHINE LEARNING BASED IMAGE CLASSIFIER FOR SHOE BRAND RECOGNITION USING CNN
What are the
metamorphic relations
in this use case ?
Train & Test whether the CNN algorithm is able to detect the image as a shoe and the brand
• Test Case 1: Happy Path Testing – Subset of the images used for training the algorithm
• Test Case 2: Metamorphic Relation 1 - Input image as a rotated image of the shoe
• Test Case 3: Metamorphic Relation 2 - Input image with BGR (instead of RGB)
• Test Case 4: Testing with the data outside of the trained data set
#ATAGTR2019
As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial
channels(Provided due credit is given to me/us)
16
Use Case & Demo
#ATAGTR2019
As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial
channels(Provided due credit is given to me/us)
17
2. Testing For Metamorphic Relations – Testing for Unstructured Data
Continuous Information flow Big Data Software Actionable Intelligence
Sentiment
Analysis
L1 - Supervised
L2 - Semi
L3 - Unsupervised
Use Case: Twitter Sentiment Analysis ML software
Vocabulary
Synonyms
Antonyms
Negations
Test case #1 – MR1: Classification using should result
Test case #2 – MR2: Classification using should result
Test case #3 – MR3: Classification using should result
Test case #4 – MR4: Classification using should result
Testing entails validation of continuous information flow
#ATAGTR2019
As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial
channels(Provided due credit is given to me/us)
18
3. Adversarial Scenario Testing
• Two-sample statistical hypothesis testing are conducted to determine both originate from the same
distribution
• Where there is high variation in distribution concludes influence of Adversial Inputs
Adversarial examples are inputs to machine learning
models that an attacker has intentionally designed to cause
the model to make a mistake
Adversarial Example:
Adversarial Input + ML Model = High accuracy (>80%) or 150% more than
normal ML model predictions
Statistical Hypothesis Testing
Attackers could target
autonomous vehicles
by using stickers or
paint to create an
adversarial stop sign
that the vehicle would
interpret as a ‘yield’ or
other sign
USE CASE
#ATAGTR2019
As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial
channels(Provided due credit is given to me/us)
19
4. Dual Coding Testing for Machine Learning Applications & Models
DCT to validate Correctness of predictions by ML applications & models
Dataset
ML Model 1 ML Model 2
Trained with same or
common set of features
Compare predictions
for correctness
Prediction 1
using RF
Prediction 2
using SVM
Random Forest
Support Vector
Machine
USE CASELEARNING MANAGEMENT SYSTEM - VDI MACHINE PREDICTION
AI enabled LMS which has to predict & allocate VDI machines based on various factors including Course,# of Course Completed, # of pending course,
Machine required. Test Data is passed through the 2 ML models namely Support Vector Machine & Random Forest.
Random Forest consistently reported higher levels of accuracy and was chosen as the base model for implementation of LMS.
#ATAGTR2019
As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial
channels(Provided due credit is given to me/us)
20
Best Practices for Testing ML Applications
ML TESTING
BEST
PRACTICES
Acceptance criteria, with
amount of error stakeholders
are willing to accept
Determine the level of outcomes
acceptable for each scenario
Communicate the level of
confidence in the results
Test with new data post
training
Scenario classification -
Expected best case, Average
case, worst case
Defects will be reflected in
the inability of the model to
achieve the goals
#ATAGTR2019
As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial
channels(Provided due credit is given to me/us)
21
Goodbye GUI, Hello VUI
• Voice First Device - Smart device designed to get tasks done conversationally
• Always-ON and Always-listening - VPA enabled Wireless Speakers
Analyst Corner (Gartner):
• By 2021, early adopter brands that redesign their websites to support visual and voice search will increase digital
commerce revenue by 30%
• By 2019, half of major commerce companies and retailers with online stores will have redesigned their commerce
sites to accommodate voice searches and voice navigation
Voice Applications
What are the Challenges in testing Voice applications? - Graphical User Interface (GUI) - Voice User Interface (VUI)
 Prescriptive vs Descriptive
 Limited data input vs indefinite # of test cases
 Testing for languages vs Testing for languages & accents
#ATAGTR2019
As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial
channels(Provided due credit is given to me/us)
22
Testing the Voice User Interface
Alexa Skill Test FrameworkTools & Frameworks Service Simulator Echosim.io Skill metrics
“Alexa, Open
Domino’s?”
“. . . ”
User
Alexa enabled Devices
Amazon – Echo, Dot, Tap, Fire TV
Triby, Pebble Watch
Alexa Voice Service (AVS)
Alexa Skills Kit
(Custom)
Services
Request Voice –
device sends to AVS
Response Voice
Request HTTPS/JSON –
POST request to Skill End point
Response
HTTPS/JSON
Speech-to-Text technology
AUT
Alexa Skills Kit - create commerce driven skill
Use Case – Leading Pizza retail chain using AI Voice Application
Avoid giving word clues, Avoid think-
aloud techniques, Avoid accidental
wake-ups
Certifies application functionality as per
design specifications in all possible
pathway and prompts
Holistic, experimental review of a
speech application; Test fully developed
application through pre-determined use-
cases
Alexa Skills Kit (ASK) - Collection of self-service APIs, tools & more. Enables designers, developers & brands to build engaging skills & reach
customers through tens of millions of Alexa Enabled Devices (AED).
Dialog Traversal Testing (DTT) VUI Review Testing (VRT) Usability Inspection Testing
#ATAGTR2019
As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial
channels(Provided due credit is given to me/us)
23
Testing the Voice User Interface – Automated Voice Testing (AVT)
Test Case
Start
Next
End
Text to Alexa Text-to-speech
Speech-to-text
Text-to-speech
Text from Alexa
Text to Alexa
“Alexa, open Domino’s”
“Welcome to Domino’s, what would
you like to order?”
“. . .”
Speech Recognition Engine
Echo
 Codeless Automation tool - TestArchitect – Extend to C#, Java or Python
 Based on Speech Recognition Engine (SRE)
 Action-based, Interaction data table
• Device Agnostic and platform agnostic
• AVT can be applied to Alexa apps, Google Home apps, voice enabled web
sites
Features Benefits
Sample Utterance
• Variations of the sample utterances with different slot values and slightly different phrases
• Slots for user errors: Test different permutations, such as missing or incorrect values
• Provide multi accent intent, verify its capabilities to process the required ‘Intents’ & respond with the appropriate ‘Utterances’
Wake Word
Alexa,
Intent Slot
To Plan my trip, For next Friday
Test Scenarios
Invocation name
Ask Yatra,
#ATAGTR2019
As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial
channels(Provided due credit is given to me/us)
24
Automated Voice Testing (AVT) Demo
#ATAGTR2019
As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial
channels(Provided due credit is given to me/us)
25
Testing Strategy for Other Areas - Chatbots, AR/VR
#ATAGTR2019
As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial
channels(Provided due credit is given to me/us)
26
NLP : Chatbot Overview & Challenges
A computer program that simulates human conversation through voice commands or text chats
or both.
What are the Challenges in testing a Chatbot?
 Unstructured Input - There are no barriers for users in terms of asking questions
 Multitude of user interactions - 100% of Test coverage is not possible
 Non-deterministic Behavior - behavior keeps changing based on learnings
What is a Chatbot ?
#ATAGTR2019
As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial
channels(Provided due credit is given to me/us)
27
Chatbot Testing
• Conversational flow - Chatbots are based on conversations &
very important to test conversational flow of a Chabot
• Understanding - Process any kind of messages
• Error Management – how does a chatbot react to errors
• Conversational steps - minimize the number of steps in
conversation
• Bot speed - Speed at which the bot
replies to the messages.
• Bot Accuracy - Out of predicted
utterances, # of utterances the bot gets
correct will be accuracy of bot.
Tools & Frameworks
Functional Non Functional
USE CASES
Travel & Hospitality
Bots can help customers plan and book trips, push personalized offers based on browsing history and preferences
Content Distribution and News
Push personalized news content, manage polls, unleash predictive content delivery based on behavior trends
Health Care
Patients leverage bots for appointments , receive personal updates based on treatment history etc.
#ATAGTR2019
As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial
channels(Provided due credit is given to me/us)
28
AR/VR Applications and Testing Challenges
• Augmented Reality - Augments virtual elements into the real world
• Virtual Reality – Digital creation of virtual environment around the user
• AR/VR has a market potential of $95B by 2025
AR / VR Applications
What are the Challenges in testing AR/VR applications?
 Wide varieties of device conditions to be dealt with and Multiple compatibility Issues
 High performing system – Performance challenges to create virtual / augmented word
 Need for consistently rich user experience
 Test Automation of AR/VR systems
Magazine ads instantly shoppable - consumers
purchase straight off the page
Project their selected item (through video) in their
living room to check suitability
Alibaba offers VR shopping experience (BUY+) through
mobile & VR Headsets - users get the experience of
shopping in an actual store
#ATAGTR2019
As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial
channels(Provided due credit is given to me/us)
29
Testing Strategy for AR/VR Applications
FUNCTIONAL TESTING
USABILITY TESTING
PERFORMANCE TESTING
SYSTEM INTERACTION ASSURANCE
• Validate Navigation, interactive buttons
• Validate interaction with voice commands etc.
AUGMENTATION ASSURANCE
• Validate/Captures/ registers accurate dimensions of a 2D/3D
object
• Real time tracking
NETWORK
ASSURANCE
• Test for varied
network related
issues (Wi-
Fi/2G/3G/4G/5G )
GEO LOCATION – LOCAL AND GLOBAL
• Ensure obtains local and global coordinates
• Marker less apps integration with SLAM (simultaneous
localization and mapping)
COMPATIBILITY
TESTING
• For integrated
and/or hosting
devices and
bandwidth
EFFECTIVENESS
• Match between system
and the real world
• Visibility of system status
(real time)
• User control and freedom
EFFICIENCY
• Consistency and standards
• Error prevention
• Accuracy
• Recognition rather than
recall (learnability) , Help
users recognize, diagnose,
and recover from errors
• Environment configuration
SATISFACTION
• Aesthetic and minimalist
design
• Help and documentation
• Customer experience with
the system
LATENCY TESTING
• Test the quality of CGI and framerate for the visuals on stress
conditions
SCALABILITY TESTING
• For cloud based AR/VR systems - dynamically scale the application
resource to support increased demand
PERFORMANCE UNDER DIFFERENT INPUTS
• Assure performance for change in angle , orientation, distance
between AR image and smart device/smart phone etc.
#ATAGTR2019
As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial
channels(Provided due credit is given to me/us)
30
AR/VR Automation
AR/VR automation can be possible by building tools with latest technologies
Computer
Vision
Robotics Device technologies will be
integrated to Automation tools automate
AR/VR Systems that functions based
motion detection, to simulate user
interactions
AI based Computer vision libraries
are going to help Automation
Tools/Bots understand the visuals /
response from AR/VR systems to
automate the interactions.
Robotic Motion
Simulator
Natural Language processing will be
another Key element for the automation
tools/bot to handle; AR/VR systems that
accepts voice commands or allow voice
interactions
Natural Language
Processing
AI based bots that can monitor and learn human interactions with systems captured from
Functional / Crowd / Beta testing and replicate for regression automation. Following are the
key technologies that will allow the Tool to achieve AR/VR automation
#ATAGTR2019
As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial
channels(Provided due credit is given to me/us)
31
In Summary
1
2
3
4
1 . AI I S N E AR I N G
M AI N S T R E A M AD O P T I O N
AI has already picked up pace across various industry
sectors with the aim of deriving business value in terms
of enhanced customer experience, new sources of
revenue and cost reduction
2 . T E S T I N G O F AI AP P S I S A
C H AL L E N G E
Non-deterministic and probabilistic nature, ever-changing
behavior based on learnings and non-linear inputs (e.g.
voice, conversational text, images/videos etc.).
3 . R O B U S T S T R AT E G Y TO
AS S U R E AI
Machine Learning Models (Feature Testing,
Metamorphic Testing, Adversarial Scenario testing,
Dual Coded ML etc.), Voice User Interfaces Testing and
Chatbots Testing
4 . M E AS U R E T H R O U G H
L E V E L O F C O N F I D E N C E
Achieve the defined acceptance criteria and
confidence levels to make the application available
for wide scale adoption

Mais conteúdo relacionado

Semelhante a #ATAGTR2019 Presentation "Assuring Quality for AI based applications" By Vinod Sundararaju Antony, Senthilkumar Thirumalaisamy & Santhosh Kumar Vasudevan

#ATAGTR2018 Presentation "Machine Learning as a decision support system for Q...
#ATAGTR2018 Presentation "Machine Learning as a decision support system for Q...#ATAGTR2018 Presentation "Machine Learning as a decision support system for Q...
#ATAGTR2018 Presentation "Machine Learning as a decision support system for Q...Agile Testing Alliance
 
#ATAGTR2020 Presentation - Non-Functional Testing of Chatbots
#ATAGTR2020 Presentation - Non-Functional Testing of Chatbots#ATAGTR2020 Presentation - Non-Functional Testing of Chatbots
#ATAGTR2020 Presentation - Non-Functional Testing of ChatbotsAgile Testing Alliance
 
#ATAGTR2018 Presentation "Design Patterns in Test Automation" By Shrinathacha...
#ATAGTR2018 Presentation "Design Patterns in Test Automation" By Shrinathacha...#ATAGTR2018 Presentation "Design Patterns in Test Automation" By Shrinathacha...
#ATAGTR2018 Presentation "Design Patterns in Test Automation" By Shrinathacha...Agile Testing Alliance
 
#ATAGTR2019 Presentation "Security testing using ML(Machine learning), AI(Art...
#ATAGTR2019 Presentation "Security testing using ML(Machine learning), AI(Art...#ATAGTR2019 Presentation "Security testing using ML(Machine learning), AI(Art...
#ATAGTR2019 Presentation "Security testing using ML(Machine learning), AI(Art...Agile Testing Alliance
 
#ATAGTR2019 Presentation "Curtailing Automation Impediments: Refactoring Auto...
#ATAGTR2019 Presentation "Curtailing Automation Impediments: Refactoring Auto...#ATAGTR2019 Presentation "Curtailing Automation Impediments: Refactoring Auto...
#ATAGTR2019 Presentation "Curtailing Automation Impediments: Refactoring Auto...Agile Testing Alliance
 
#ATAGTR2020 Presentation - Adaptive Learner
#ATAGTR2020 Presentation - Adaptive Learner#ATAGTR2020 Presentation - Adaptive Learner
#ATAGTR2020 Presentation - Adaptive LearnerAgile Testing Alliance
 
#ATAGTR2019 Presentation "Re-imagining Quality Assurance in the age of Digiti...
#ATAGTR2019 Presentation "Re-imagining Quality Assurance in the age of Digiti...#ATAGTR2019 Presentation "Re-imagining Quality Assurance in the age of Digiti...
#ATAGTR2019 Presentation "Re-imagining Quality Assurance in the age of Digiti...Agile Testing Alliance
 
#ATAGTR2019 Presentation "How analytics is reshaping testing?!" By Himani Yad...
#ATAGTR2019 Presentation "How analytics is reshaping testing?!" By Himani Yad...#ATAGTR2019 Presentation "How analytics is reshaping testing?!" By Himani Yad...
#ATAGTR2019 Presentation "How analytics is reshaping testing?!" By Himani Yad...Agile Testing Alliance
 
Demonetization, IoT and related thoughts!
Demonetization, IoT and related thoughts!Demonetization, IoT and related thoughts!
Demonetization, IoT and related thoughts!Agile Testing Alliance
 
#ATAGTR2019 Presentation "Swallowables – The Next QA Frontier?" By Jibendu Na...
#ATAGTR2019 Presentation "Swallowables – The Next QA Frontier?" By Jibendu Na...#ATAGTR2019 Presentation "Swallowables – The Next QA Frontier?" By Jibendu Na...
#ATAGTR2019 Presentation "Swallowables – The Next QA Frontier?" By Jibendu Na...Agile Testing Alliance
 
#ATAGTR2020 Presentation - Redefining DevOps for seamless performance testing
#ATAGTR2020 Presentation - Redefining DevOps for seamless performance testing#ATAGTR2020 Presentation - Redefining DevOps for seamless performance testing
#ATAGTR2020 Presentation - Redefining DevOps for seamless performance testingAgile Testing Alliance
 
#ATAGTR2020 Presentation - The Splunk Integration for Futuristic NFT in DevOp...
#ATAGTR2020 Presentation - The Splunk Integration for Futuristic NFT in DevOp...#ATAGTR2020 Presentation - The Splunk Integration for Futuristic NFT in DevOp...
#ATAGTR2020 Presentation - The Splunk Integration for Futuristic NFT in DevOp...Agile Testing Alliance
 
Addressing the challenges of delivering Microservice applications in the ente...
Addressing the challenges of delivering Microservice applications in the ente...Addressing the challenges of delivering Microservice applications in the ente...
Addressing the challenges of delivering Microservice applications in the ente...Agile Testing Alliance
 
The New Categories of Software Defects in the Era of AI and ML - DevOps Next
The New Categories of Software Defects in the Era of AI and ML - DevOps NextThe New Categories of Software Defects in the Era of AI and ML - DevOps Next
The New Categories of Software Defects in the Era of AI and ML - DevOps NextPerfecto by Perforce
 
#ATAGTR2018 Presentation "How a fortune 500 case Study Changed The Way We Do ...
#ATAGTR2018 Presentation "How a fortune 500 case Study Changed The Way We Do ...#ATAGTR2018 Presentation "How a fortune 500 case Study Changed The Way We Do ...
#ATAGTR2018 Presentation "How a fortune 500 case Study Changed The Way We Do ...Agile Testing Alliance
 
Building Your Own AI Instance (TBLC AI )
Building Your Own AI Instance (TBLC AI )Building Your Own AI Instance (TBLC AI )
Building Your Own AI Instance (TBLC AI )Brian Pichman
 
#ATAGTR2018 Presentation " Security Testing for RESTful APIs" By Anuradha Raman
#ATAGTR2018 Presentation " Security Testing for RESTful APIs" By Anuradha Raman #ATAGTR2018 Presentation " Security Testing for RESTful APIs" By Anuradha Raman
#ATAGTR2018 Presentation " Security Testing for RESTful APIs" By Anuradha Raman Agile Testing Alliance
 
#ATAGTR2020 Presentation - Relish your journey to Software Testing Masterchef
#ATAGTR2020 Presentation - Relish your journey to Software Testing Masterchef#ATAGTR2020 Presentation - Relish your journey to Software Testing Masterchef
#ATAGTR2020 Presentation - Relish your journey to Software Testing MasterchefAgile Testing Alliance
 
#ATAGTR2020 Presentation - Case study for holistic approach to IoT testing
#ATAGTR2020 Presentation - Case study for holistic approach to IoT testing#ATAGTR2020 Presentation - Case study for holistic approach to IoT testing
#ATAGTR2020 Presentation - Case study for holistic approach to IoT testingAgile Testing Alliance
 

Semelhante a #ATAGTR2019 Presentation "Assuring Quality for AI based applications" By Vinod Sundararaju Antony, Senthilkumar Thirumalaisamy & Santhosh Kumar Vasudevan (20)

#ATAGTR2018 Presentation "Machine Learning as a decision support system for Q...
#ATAGTR2018 Presentation "Machine Learning as a decision support system for Q...#ATAGTR2018 Presentation "Machine Learning as a decision support system for Q...
#ATAGTR2018 Presentation "Machine Learning as a decision support system for Q...
 
#ATAGTR2020 Presentation - Non-Functional Testing of Chatbots
#ATAGTR2020 Presentation - Non-Functional Testing of Chatbots#ATAGTR2020 Presentation - Non-Functional Testing of Chatbots
#ATAGTR2020 Presentation - Non-Functional Testing of Chatbots
 
#ATAGTR2018 Presentation "Design Patterns in Test Automation" By Shrinathacha...
#ATAGTR2018 Presentation "Design Patterns in Test Automation" By Shrinathacha...#ATAGTR2018 Presentation "Design Patterns in Test Automation" By Shrinathacha...
#ATAGTR2018 Presentation "Design Patterns in Test Automation" By Shrinathacha...
 
#ATAGTR2019 Presentation "Security testing using ML(Machine learning), AI(Art...
#ATAGTR2019 Presentation "Security testing using ML(Machine learning), AI(Art...#ATAGTR2019 Presentation "Security testing using ML(Machine learning), AI(Art...
#ATAGTR2019 Presentation "Security testing using ML(Machine learning), AI(Art...
 
#ATAGTR2019 Presentation "Curtailing Automation Impediments: Refactoring Auto...
#ATAGTR2019 Presentation "Curtailing Automation Impediments: Refactoring Auto...#ATAGTR2019 Presentation "Curtailing Automation Impediments: Refactoring Auto...
#ATAGTR2019 Presentation "Curtailing Automation Impediments: Refactoring Auto...
 
#ATAGTR2020 Presentation - Adaptive Learner
#ATAGTR2020 Presentation - Adaptive Learner#ATAGTR2020 Presentation - Adaptive Learner
#ATAGTR2020 Presentation - Adaptive Learner
 
#ATAGTR2019 Presentation "Re-imagining Quality Assurance in the age of Digiti...
#ATAGTR2019 Presentation "Re-imagining Quality Assurance in the age of Digiti...#ATAGTR2019 Presentation "Re-imagining Quality Assurance in the age of Digiti...
#ATAGTR2019 Presentation "Re-imagining Quality Assurance in the age of Digiti...
 
#ATAGTR2019 Presentation "How analytics is reshaping testing?!" By Himani Yad...
#ATAGTR2019 Presentation "How analytics is reshaping testing?!" By Himani Yad...#ATAGTR2019 Presentation "How analytics is reshaping testing?!" By Himani Yad...
#ATAGTR2019 Presentation "How analytics is reshaping testing?!" By Himani Yad...
 
Demonetization, IoT and related thoughts!
Demonetization, IoT and related thoughts!Demonetization, IoT and related thoughts!
Demonetization, IoT and related thoughts!
 
#ATAGTR2019 Presentation "Swallowables – The Next QA Frontier?" By Jibendu Na...
#ATAGTR2019 Presentation "Swallowables – The Next QA Frontier?" By Jibendu Na...#ATAGTR2019 Presentation "Swallowables – The Next QA Frontier?" By Jibendu Na...
#ATAGTR2019 Presentation "Swallowables – The Next QA Frontier?" By Jibendu Na...
 
#ATAGTR2020 Presentation - Redefining DevOps for seamless performance testing
#ATAGTR2020 Presentation - Redefining DevOps for seamless performance testing#ATAGTR2020 Presentation - Redefining DevOps for seamless performance testing
#ATAGTR2020 Presentation - Redefining DevOps for seamless performance testing
 
#ATAGTR2020 Presentation - The Splunk Integration for Futuristic NFT in DevOp...
#ATAGTR2020 Presentation - The Splunk Integration for Futuristic NFT in DevOp...#ATAGTR2020 Presentation - The Splunk Integration for Futuristic NFT in DevOp...
#ATAGTR2020 Presentation - The Splunk Integration for Futuristic NFT in DevOp...
 
Addressing the challenges of delivering Microservice applications in the ente...
Addressing the challenges of delivering Microservice applications in the ente...Addressing the challenges of delivering Microservice applications in the ente...
Addressing the challenges of delivering Microservice applications in the ente...
 
The New Categories of Software Defects in the Era of AI and ML - DevOps Next
The New Categories of Software Defects in the Era of AI and ML - DevOps NextThe New Categories of Software Defects in the Era of AI and ML - DevOps Next
The New Categories of Software Defects in the Era of AI and ML - DevOps Next
 
#ATAGTR2018 Presentation "How a fortune 500 case Study Changed The Way We Do ...
#ATAGTR2018 Presentation "How a fortune 500 case Study Changed The Way We Do ...#ATAGTR2018 Presentation "How a fortune 500 case Study Changed The Way We Do ...
#ATAGTR2018 Presentation "How a fortune 500 case Study Changed The Way We Do ...
 
Building Your Own AI Instance (TBLC AI )
Building Your Own AI Instance (TBLC AI )Building Your Own AI Instance (TBLC AI )
Building Your Own AI Instance (TBLC AI )
 
EIS-Webinar- Generative-AI-KM-2023-04-19.pdf
EIS-Webinar- Generative-AI-KM-2023-04-19.pdfEIS-Webinar- Generative-AI-KM-2023-04-19.pdf
EIS-Webinar- Generative-AI-KM-2023-04-19.pdf
 
#ATAGTR2018 Presentation " Security Testing for RESTful APIs" By Anuradha Raman
#ATAGTR2018 Presentation " Security Testing for RESTful APIs" By Anuradha Raman #ATAGTR2018 Presentation " Security Testing for RESTful APIs" By Anuradha Raman
#ATAGTR2018 Presentation " Security Testing for RESTful APIs" By Anuradha Raman
 
#ATAGTR2020 Presentation - Relish your journey to Software Testing Masterchef
#ATAGTR2020 Presentation - Relish your journey to Software Testing Masterchef#ATAGTR2020 Presentation - Relish your journey to Software Testing Masterchef
#ATAGTR2020 Presentation - Relish your journey to Software Testing Masterchef
 
#ATAGTR2020 Presentation - Case study for holistic approach to IoT testing
#ATAGTR2020 Presentation - Case study for holistic approach to IoT testing#ATAGTR2020 Presentation - Case study for holistic approach to IoT testing
#ATAGTR2020 Presentation - Case study for holistic approach to IoT testing
 

Mais de Agile Testing Alliance

#Interactive Session by Anindita Rath and Mahathee Dandibhotla, "From Good to...
#Interactive Session by Anindita Rath and Mahathee Dandibhotla, "From Good to...#Interactive Session by Anindita Rath and Mahathee Dandibhotla, "From Good to...
#Interactive Session by Anindita Rath and Mahathee Dandibhotla, "From Good to...Agile Testing Alliance
 
#Interactive Session by Ajay Balamurugadas, "Where Are The Real Testers In T...
#Interactive Session by  Ajay Balamurugadas, "Where Are The Real Testers In T...#Interactive Session by  Ajay Balamurugadas, "Where Are The Real Testers In T...
#Interactive Session by Ajay Balamurugadas, "Where Are The Real Testers In T...Agile Testing Alliance
 
#Interactive Session by Jishnu Nambiar and Mayur Ovhal, "Monitoring Web Per...
#Interactive Session by  Jishnu Nambiar and  Mayur Ovhal, "Monitoring Web Per...#Interactive Session by  Jishnu Nambiar and  Mayur Ovhal, "Monitoring Web Per...
#Interactive Session by Jishnu Nambiar and Mayur Ovhal, "Monitoring Web Per...Agile Testing Alliance
 
#Interactive Session by Pradipta Biswas and Sucheta Saurabh Chitale, "Navigat...
#Interactive Session by Pradipta Biswas and Sucheta Saurabh Chitale, "Navigat...#Interactive Session by Pradipta Biswas and Sucheta Saurabh Chitale, "Navigat...
#Interactive Session by Pradipta Biswas and Sucheta Saurabh Chitale, "Navigat...Agile Testing Alliance
 
#Interactive Session by Apoorva Ram, "The Art of Storytelling for Testers" at...
#Interactive Session by Apoorva Ram, "The Art of Storytelling for Testers" at...#Interactive Session by Apoorva Ram, "The Art of Storytelling for Testers" at...
#Interactive Session by Apoorva Ram, "The Art of Storytelling for Testers" at...Agile Testing Alliance
 
#Interactive Session by Nikhil Jain, "Catch All Mail With Graph" at #ATAGTR2023.
#Interactive Session by Nikhil Jain, "Catch All Mail With Graph" at #ATAGTR2023.#Interactive Session by Nikhil Jain, "Catch All Mail With Graph" at #ATAGTR2023.
#Interactive Session by Nikhil Jain, "Catch All Mail With Graph" at #ATAGTR2023.Agile Testing Alliance
 
#Interactive Session by Ashok Kumar S, "Test Data the key to robust test cove...
#Interactive Session by Ashok Kumar S, "Test Data the key to robust test cove...#Interactive Session by Ashok Kumar S, "Test Data the key to robust test cove...
#Interactive Session by Ashok Kumar S, "Test Data the key to robust test cove...Agile Testing Alliance
 
#Interactive Session by Seema Kohli, "Test Leadership in the Era of Artificia...
#Interactive Session by Seema Kohli, "Test Leadership in the Era of Artificia...#Interactive Session by Seema Kohli, "Test Leadership in the Era of Artificia...
#Interactive Session by Seema Kohli, "Test Leadership in the Era of Artificia...Agile Testing Alliance
 
#Interactive Session by Ashwini Lalit, RRR of Test Automation Maintenance" at...
#Interactive Session by Ashwini Lalit, RRR of Test Automation Maintenance" at...#Interactive Session by Ashwini Lalit, RRR of Test Automation Maintenance" at...
#Interactive Session by Ashwini Lalit, RRR of Test Automation Maintenance" at...Agile Testing Alliance
 
#Interactive Session by Srithanga Aishvarya T, "Machine Learning Model to aut...
#Interactive Session by Srithanga Aishvarya T, "Machine Learning Model to aut...#Interactive Session by Srithanga Aishvarya T, "Machine Learning Model to aut...
#Interactive Session by Srithanga Aishvarya T, "Machine Learning Model to aut...Agile Testing Alliance
 
#Interactive Session by Kirti Ranjan Satapathy and Nandini K, "Elements of Qu...
#Interactive Session by Kirti Ranjan Satapathy and Nandini K, "Elements of Qu...#Interactive Session by Kirti Ranjan Satapathy and Nandini K, "Elements of Qu...
#Interactive Session by Kirti Ranjan Satapathy and Nandini K, "Elements of Qu...Agile Testing Alliance
 
#Interactive Session by Sudhir Upadhyay and Ashish Kumar, "Strengthening Test...
#Interactive Session by Sudhir Upadhyay and Ashish Kumar, "Strengthening Test...#Interactive Session by Sudhir Upadhyay and Ashish Kumar, "Strengthening Test...
#Interactive Session by Sudhir Upadhyay and Ashish Kumar, "Strengthening Test...Agile Testing Alliance
 
#Interactive Session by Sayan Deb Kundu, "Testing Gen AI Applications" at #AT...
#Interactive Session by Sayan Deb Kundu, "Testing Gen AI Applications" at #AT...#Interactive Session by Sayan Deb Kundu, "Testing Gen AI Applications" at #AT...
#Interactive Session by Sayan Deb Kundu, "Testing Gen AI Applications" at #AT...Agile Testing Alliance
 
#Interactive Session by Dinesh Boravke, "Zero Defects – Myth or Reality" at #...
#Interactive Session by Dinesh Boravke, "Zero Defects – Myth or Reality" at #...#Interactive Session by Dinesh Boravke, "Zero Defects – Myth or Reality" at #...
#Interactive Session by Dinesh Boravke, "Zero Defects – Myth or Reality" at #...Agile Testing Alliance
 
#Interactive Session by Saby Saurabh Bhardwaj, "Redefine Quality Assurance –...
#Interactive Session by  Saby Saurabh Bhardwaj, "Redefine Quality Assurance –...#Interactive Session by  Saby Saurabh Bhardwaj, "Redefine Quality Assurance –...
#Interactive Session by Saby Saurabh Bhardwaj, "Redefine Quality Assurance –...Agile Testing Alliance
 
#Keynote Session by Sanjay Kumar, "Innovation Inspired Testing!!" at #ATAGTR2...
#Keynote Session by Sanjay Kumar, "Innovation Inspired Testing!!" at #ATAGTR2...#Keynote Session by Sanjay Kumar, "Innovation Inspired Testing!!" at #ATAGTR2...
#Keynote Session by Sanjay Kumar, "Innovation Inspired Testing!!" at #ATAGTR2...Agile Testing Alliance
 
#Keynote Session by Schalk Cronje, "Don’t Containerize me" at #ATAGTR2023.
#Keynote Session by Schalk Cronje, "Don’t Containerize me" at #ATAGTR2023.#Keynote Session by Schalk Cronje, "Don’t Containerize me" at #ATAGTR2023.
#Keynote Session by Schalk Cronje, "Don’t Containerize me" at #ATAGTR2023.Agile Testing Alliance
 
#Interactive Session by Chidambaram Vetrivel and Venkatesh Belde, "Revolution...
#Interactive Session by Chidambaram Vetrivel and Venkatesh Belde, "Revolution...#Interactive Session by Chidambaram Vetrivel and Venkatesh Belde, "Revolution...
#Interactive Session by Chidambaram Vetrivel and Venkatesh Belde, "Revolution...Agile Testing Alliance
 
#Interactive Session by Aniket Diwakar Kadukar and Padimiti Vaidik Eswar Dat...
#Interactive Session by Aniket Diwakar Kadukar and  Padimiti Vaidik Eswar Dat...#Interactive Session by Aniket Diwakar Kadukar and  Padimiti Vaidik Eswar Dat...
#Interactive Session by Aniket Diwakar Kadukar and Padimiti Vaidik Eswar Dat...Agile Testing Alliance
 
#Interactive Session by Vivek Patle and Jahnavi Umarji, "Empowering Functiona...
#Interactive Session by Vivek Patle and Jahnavi Umarji, "Empowering Functiona...#Interactive Session by Vivek Patle and Jahnavi Umarji, "Empowering Functiona...
#Interactive Session by Vivek Patle and Jahnavi Umarji, "Empowering Functiona...Agile Testing Alliance
 

Mais de Agile Testing Alliance (20)

#Interactive Session by Anindita Rath and Mahathee Dandibhotla, "From Good to...
#Interactive Session by Anindita Rath and Mahathee Dandibhotla, "From Good to...#Interactive Session by Anindita Rath and Mahathee Dandibhotla, "From Good to...
#Interactive Session by Anindita Rath and Mahathee Dandibhotla, "From Good to...
 
#Interactive Session by Ajay Balamurugadas, "Where Are The Real Testers In T...
#Interactive Session by  Ajay Balamurugadas, "Where Are The Real Testers In T...#Interactive Session by  Ajay Balamurugadas, "Where Are The Real Testers In T...
#Interactive Session by Ajay Balamurugadas, "Where Are The Real Testers In T...
 
#Interactive Session by Jishnu Nambiar and Mayur Ovhal, "Monitoring Web Per...
#Interactive Session by  Jishnu Nambiar and  Mayur Ovhal, "Monitoring Web Per...#Interactive Session by  Jishnu Nambiar and  Mayur Ovhal, "Monitoring Web Per...
#Interactive Session by Jishnu Nambiar and Mayur Ovhal, "Monitoring Web Per...
 
#Interactive Session by Pradipta Biswas and Sucheta Saurabh Chitale, "Navigat...
#Interactive Session by Pradipta Biswas and Sucheta Saurabh Chitale, "Navigat...#Interactive Session by Pradipta Biswas and Sucheta Saurabh Chitale, "Navigat...
#Interactive Session by Pradipta Biswas and Sucheta Saurabh Chitale, "Navigat...
 
#Interactive Session by Apoorva Ram, "The Art of Storytelling for Testers" at...
#Interactive Session by Apoorva Ram, "The Art of Storytelling for Testers" at...#Interactive Session by Apoorva Ram, "The Art of Storytelling for Testers" at...
#Interactive Session by Apoorva Ram, "The Art of Storytelling for Testers" at...
 
#Interactive Session by Nikhil Jain, "Catch All Mail With Graph" at #ATAGTR2023.
#Interactive Session by Nikhil Jain, "Catch All Mail With Graph" at #ATAGTR2023.#Interactive Session by Nikhil Jain, "Catch All Mail With Graph" at #ATAGTR2023.
#Interactive Session by Nikhil Jain, "Catch All Mail With Graph" at #ATAGTR2023.
 
#Interactive Session by Ashok Kumar S, "Test Data the key to robust test cove...
#Interactive Session by Ashok Kumar S, "Test Data the key to robust test cove...#Interactive Session by Ashok Kumar S, "Test Data the key to robust test cove...
#Interactive Session by Ashok Kumar S, "Test Data the key to robust test cove...
 
#Interactive Session by Seema Kohli, "Test Leadership in the Era of Artificia...
#Interactive Session by Seema Kohli, "Test Leadership in the Era of Artificia...#Interactive Session by Seema Kohli, "Test Leadership in the Era of Artificia...
#Interactive Session by Seema Kohli, "Test Leadership in the Era of Artificia...
 
#Interactive Session by Ashwini Lalit, RRR of Test Automation Maintenance" at...
#Interactive Session by Ashwini Lalit, RRR of Test Automation Maintenance" at...#Interactive Session by Ashwini Lalit, RRR of Test Automation Maintenance" at...
#Interactive Session by Ashwini Lalit, RRR of Test Automation Maintenance" at...
 
#Interactive Session by Srithanga Aishvarya T, "Machine Learning Model to aut...
#Interactive Session by Srithanga Aishvarya T, "Machine Learning Model to aut...#Interactive Session by Srithanga Aishvarya T, "Machine Learning Model to aut...
#Interactive Session by Srithanga Aishvarya T, "Machine Learning Model to aut...
 
#Interactive Session by Kirti Ranjan Satapathy and Nandini K, "Elements of Qu...
#Interactive Session by Kirti Ranjan Satapathy and Nandini K, "Elements of Qu...#Interactive Session by Kirti Ranjan Satapathy and Nandini K, "Elements of Qu...
#Interactive Session by Kirti Ranjan Satapathy and Nandini K, "Elements of Qu...
 
#Interactive Session by Sudhir Upadhyay and Ashish Kumar, "Strengthening Test...
#Interactive Session by Sudhir Upadhyay and Ashish Kumar, "Strengthening Test...#Interactive Session by Sudhir Upadhyay and Ashish Kumar, "Strengthening Test...
#Interactive Session by Sudhir Upadhyay and Ashish Kumar, "Strengthening Test...
 
#Interactive Session by Sayan Deb Kundu, "Testing Gen AI Applications" at #AT...
#Interactive Session by Sayan Deb Kundu, "Testing Gen AI Applications" at #AT...#Interactive Session by Sayan Deb Kundu, "Testing Gen AI Applications" at #AT...
#Interactive Session by Sayan Deb Kundu, "Testing Gen AI Applications" at #AT...
 
#Interactive Session by Dinesh Boravke, "Zero Defects – Myth or Reality" at #...
#Interactive Session by Dinesh Boravke, "Zero Defects – Myth or Reality" at #...#Interactive Session by Dinesh Boravke, "Zero Defects – Myth or Reality" at #...
#Interactive Session by Dinesh Boravke, "Zero Defects – Myth or Reality" at #...
 
#Interactive Session by Saby Saurabh Bhardwaj, "Redefine Quality Assurance –...
#Interactive Session by  Saby Saurabh Bhardwaj, "Redefine Quality Assurance –...#Interactive Session by  Saby Saurabh Bhardwaj, "Redefine Quality Assurance –...
#Interactive Session by Saby Saurabh Bhardwaj, "Redefine Quality Assurance –...
 
#Keynote Session by Sanjay Kumar, "Innovation Inspired Testing!!" at #ATAGTR2...
#Keynote Session by Sanjay Kumar, "Innovation Inspired Testing!!" at #ATAGTR2...#Keynote Session by Sanjay Kumar, "Innovation Inspired Testing!!" at #ATAGTR2...
#Keynote Session by Sanjay Kumar, "Innovation Inspired Testing!!" at #ATAGTR2...
 
#Keynote Session by Schalk Cronje, "Don’t Containerize me" at #ATAGTR2023.
#Keynote Session by Schalk Cronje, "Don’t Containerize me" at #ATAGTR2023.#Keynote Session by Schalk Cronje, "Don’t Containerize me" at #ATAGTR2023.
#Keynote Session by Schalk Cronje, "Don’t Containerize me" at #ATAGTR2023.
 
#Interactive Session by Chidambaram Vetrivel and Venkatesh Belde, "Revolution...
#Interactive Session by Chidambaram Vetrivel and Venkatesh Belde, "Revolution...#Interactive Session by Chidambaram Vetrivel and Venkatesh Belde, "Revolution...
#Interactive Session by Chidambaram Vetrivel and Venkatesh Belde, "Revolution...
 
#Interactive Session by Aniket Diwakar Kadukar and Padimiti Vaidik Eswar Dat...
#Interactive Session by Aniket Diwakar Kadukar and  Padimiti Vaidik Eswar Dat...#Interactive Session by Aniket Diwakar Kadukar and  Padimiti Vaidik Eswar Dat...
#Interactive Session by Aniket Diwakar Kadukar and Padimiti Vaidik Eswar Dat...
 
#Interactive Session by Vivek Patle and Jahnavi Umarji, "Empowering Functiona...
#Interactive Session by Vivek Patle and Jahnavi Umarji, "Empowering Functiona...#Interactive Session by Vivek Patle and Jahnavi Umarji, "Empowering Functiona...
#Interactive Session by Vivek Patle and Jahnavi Umarji, "Empowering Functiona...
 

Último

Histor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slideHistor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slidevu2urc
 
Powerful Google developer tools for immediate impact! (2023-24 C)
Powerful Google developer tools for immediate impact! (2023-24 C)Powerful Google developer tools for immediate impact! (2023-24 C)
Powerful Google developer tools for immediate impact! (2023-24 C)wesley chun
 
What Are The Drone Anti-jamming Systems Technology?
What Are The Drone Anti-jamming Systems Technology?What Are The Drone Anti-jamming Systems Technology?
What Are The Drone Anti-jamming Systems Technology?Antenna Manufacturer Coco
 
Automating Google Workspace (GWS) & more with Apps Script
Automating Google Workspace (GWS) & more with Apps ScriptAutomating Google Workspace (GWS) & more with Apps Script
Automating Google Workspace (GWS) & more with Apps Scriptwesley chun
 
Presentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreterPresentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreternaman860154
 
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024The Digital Insurer
 
🐬 The future of MySQL is Postgres 🐘
🐬  The future of MySQL is Postgres   🐘🐬  The future of MySQL is Postgres   🐘
🐬 The future of MySQL is Postgres 🐘RTylerCroy
 
Scaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organizationScaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organizationRadu Cotescu
 
CNv6 Instructor Chapter 6 Quality of Service
CNv6 Instructor Chapter 6 Quality of ServiceCNv6 Instructor Chapter 6 Quality of Service
CNv6 Instructor Chapter 6 Quality of Servicegiselly40
 
A Call to Action for Generative AI in 2024
A Call to Action for Generative AI in 2024A Call to Action for Generative AI in 2024
A Call to Action for Generative AI in 2024Results
 
2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...Martijn de Jong
 
Axa Assurance Maroc - Insurer Innovation Award 2024
Axa Assurance Maroc - Insurer Innovation Award 2024Axa Assurance Maroc - Insurer Innovation Award 2024
Axa Assurance Maroc - Insurer Innovation Award 2024The Digital Insurer
 
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...apidays
 
Factors to Consider When Choosing Accounts Payable Services Providers.pptx
Factors to Consider When Choosing Accounts Payable Services Providers.pptxFactors to Consider When Choosing Accounts Payable Services Providers.pptx
Factors to Consider When Choosing Accounts Payable Services Providers.pptxKatpro Technologies
 
IAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI SolutionsIAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI SolutionsEnterprise Knowledge
 
Real Time Object Detection Using Open CV
Real Time Object Detection Using Open CVReal Time Object Detection Using Open CV
Real Time Object Detection Using Open CVKhem
 
The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024Rafal Los
 
Boost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivityBoost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivityPrincipled Technologies
 
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking MenDelhi Call girls
 
Boost Fertility New Invention Ups Success Rates.pdf
Boost Fertility New Invention Ups Success Rates.pdfBoost Fertility New Invention Ups Success Rates.pdf
Boost Fertility New Invention Ups Success Rates.pdfsudhanshuwaghmare1
 

Último (20)

Histor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slideHistor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slide
 
Powerful Google developer tools for immediate impact! (2023-24 C)
Powerful Google developer tools for immediate impact! (2023-24 C)Powerful Google developer tools for immediate impact! (2023-24 C)
Powerful Google developer tools for immediate impact! (2023-24 C)
 
What Are The Drone Anti-jamming Systems Technology?
What Are The Drone Anti-jamming Systems Technology?What Are The Drone Anti-jamming Systems Technology?
What Are The Drone Anti-jamming Systems Technology?
 
Automating Google Workspace (GWS) & more with Apps Script
Automating Google Workspace (GWS) & more with Apps ScriptAutomating Google Workspace (GWS) & more with Apps Script
Automating Google Workspace (GWS) & more with Apps Script
 
Presentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreterPresentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreter
 
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
 
🐬 The future of MySQL is Postgres 🐘
🐬  The future of MySQL is Postgres   🐘🐬  The future of MySQL is Postgres   🐘
🐬 The future of MySQL is Postgres 🐘
 
Scaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organizationScaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organization
 
CNv6 Instructor Chapter 6 Quality of Service
CNv6 Instructor Chapter 6 Quality of ServiceCNv6 Instructor Chapter 6 Quality of Service
CNv6 Instructor Chapter 6 Quality of Service
 
A Call to Action for Generative AI in 2024
A Call to Action for Generative AI in 2024A Call to Action for Generative AI in 2024
A Call to Action for Generative AI in 2024
 
2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...
 
Axa Assurance Maroc - Insurer Innovation Award 2024
Axa Assurance Maroc - Insurer Innovation Award 2024Axa Assurance Maroc - Insurer Innovation Award 2024
Axa Assurance Maroc - Insurer Innovation Award 2024
 
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
 
Factors to Consider When Choosing Accounts Payable Services Providers.pptx
Factors to Consider When Choosing Accounts Payable Services Providers.pptxFactors to Consider When Choosing Accounts Payable Services Providers.pptx
Factors to Consider When Choosing Accounts Payable Services Providers.pptx
 
IAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI SolutionsIAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI Solutions
 
Real Time Object Detection Using Open CV
Real Time Object Detection Using Open CVReal Time Object Detection Using Open CV
Real Time Object Detection Using Open CV
 
The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024
 
Boost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivityBoost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivity
 
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
 
Boost Fertility New Invention Ups Success Rates.pdf
Boost Fertility New Invention Ups Success Rates.pdfBoost Fertility New Invention Ups Success Rates.pdf
Boost Fertility New Invention Ups Success Rates.pdf
 

#ATAGTR2019 Presentation "Assuring Quality for AI based applications" By Vinod Sundararaju Antony, Senthilkumar Thirumalaisamy & Santhosh Kumar Vasudevan

  • 1. #ATAGTR2019 Assuring Quality for AI based applications Vinod Sundararaju Antony, Senthilkumar Thirumalaisamy, Santhosh Kumar Vasudevan 14th 15th Dec 2019
  • 2. #ATAGTR2019 As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial channels(Provided due credit is given to me/us) Abstract Key challenges include getting the right data sets for testing, predicting and test designing the expected outcome and behaviour, identifying apt algorithms AI IS NEARER TO MAINSTREAM Intelligent software and applications are omnipresent and are changing the way we engage Artificial Intelligence (AI) technology is drastically finding its way into conventional software development A Robust test strategy which increases the level of confidence in AI Apps is the need of the hour 2
  • 3. #ATAGTR2019 As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial channels(Provided due credit is given to me/us) AI Is Here To Stay Global AI Business Value to Reach $1.2 Trillion in 2018 and $3.9 trillion in 2022 AI has already picked up pace across various industry sectors BANKING & FINANCE HEALTHCARE MARKETING AUTOMOBILE Travel concierge Personalized platform for Airlines TRAVEL & CUSTOMER RELATION SMART HOME INFORMATION, MEDIA & ENTERTAINMENT VIRTUAL ASSISTANT RETAIL 3
  • 4. #ATAGTR2019 As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial channels(Provided due credit is given to me/us) AI Is Nearer To Mainstream Democratized AI – AI will become more widely available due to cloud computing, open source and the “maker” community (Deep Neural Nets is just 2-5 years from mainstream adoption) 4
  • 5. #ATAGTR2019 As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial channels(Provided due credit is given to me/us) But There Is Still A Long Way To Go Google Apologizes for Photo App’s Racist Blunder Tesla Car that crashed was running on Autopilot Microsoft silences its new AI bot Tay after Twitter users teach it racism 5
  • 6. #ATAGTR2019 As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial channels(Provided due credit is given to me/us) The spectrum of AI & the challenges in assuring Quality MACHINE LEARNING Deep Learning Predictive Analytics NATURAL LANGUAGE PROCESSING (NLP) Translation Classification & Clustering Information Extraction SPEECH Speech to Text Text to Speech Expert Systems Robotics Vision Image Recognition Machine Vision Artificial Intelligence • Non-deterministic and probabilistic - No defined input and output • Ever-changing Behaviour – AI systems are always learning • Non-Linear inputs – e.g. Voice, conversational/free-flowing text 6
  • 7. #ATAGTR2019 As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial channels(Provided due credit is given to me/us) 7  Detecting credit card fraud  Determining whether a customer is likely to switch to a competitor  Deciding when to do preventive maintenance on a factory robot What is Machine Learning? Machine Learning focuses on data-driven predictions as opposed to following strictly static program instructions Uses the patterns to predict the futureFind patterns in data
  • 8. #ATAGTR2019 As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial channels(Provided due credit is given to me/us) 8 Machine Learning – How does it work ? Machine Learning Algorithm Model Application Data Contains Patterns Finds Patterns Recognizes Patterns Supplies new data to see if it matches known patterns Why is it gaining traction ? Machine Learning requires Lots of data, Lots of compute power, Effective Machine Learning Algorithms. All of these are more available than ever with the evolution of Big Data, Cloud etc.
  • 9. #ATAGTR2019 As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial channels(Provided due credit is given to me/us) 9 What are the testing challenges? ML applications are intended to learn properties of data sets, expected output is NOT already known to users Periodic learning leading to changing behaviour over period of time Prediction of all scenarios is a time consuming process Dependency on humungous amount of data?
  • 10. #ATAGTR2019 As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial channels(Provided due credit is given to me/us) 10 Testing a Machine Learning Model Features make the most important part of a machine learning application or model. Testing features are key set of Test tasks which needs to be performed for ensuring the high performance of machine learning.
  • 11. #ATAGTR2019 As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial channels(Provided due credit is given to me/us) 11 1. Testing Features of Machine Learning models Feature Thresholds Feature Relevance Feature Relationship Feature Suitability Feature Compliance Features Unit Testing Feature Static Review Test if the feature relationship with outcome variable in terms of correlation coefficients. Test whether value of features lies between the threshold values Age of Human (Y) - Threshold - 0 (X1) to 100 (X2). Test if Y lies between X1 and X2 (or) Y > X1 Test whether the feature importance changed with respect to previous test run. Test/review the static code analysis outcome of code generating features Test/review the code coverage of the code generating features Test/review if the generated feature violates the data compliance related issues Test the feature unsuitability by testing memory usage, inference latency and more.
  • 12. #ATAGTR2019 As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial channels(Provided due credit is given to me/us) 12 2. Testing for metamorphic relations Test Case 1 X=X1 Machine Learning System Under Test Output 1 Y1 Test Case 2 X=G(X1) Output 2 Y2 Input Output Metamorphic Relations Input Relation Output Relation AN AI BASED APPLICATION TO PREDICT THE RISK OF DIABETES USE CASE Age BMI Predict Diabetes (Y/ N) 30 30 Y 40 32 N Indicates failure in Metamorphic relation • Validation of relations between the outputs of multiple inputs can help detecting defects in ML algorithm • Follow-up test cases and validation of the test results and could be fully automated.
  • 13. #ATAGTR2019 As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial channels(Provided due credit is given to me/us) 13 2. Testing For Metamorphic Relations – Orientation & Color Testing ORIENTATION COLOR • RGB Channels of an image constitute pixel values for ‘red’, ‘green’ and ‘blue’ colors • Permutation of RGB input channels helps in identification of metamorphic relations and detect failures • Core property of CNN is NOT violated by RGB permutation Original, RGB BGR CNN SUT BOATTest case #1 Convolutional Neural Network Test case #1 CNN SUT BOAT Convolutional Neural Network Test Case #2 – MR1 Test Case #3 – MR2 90o 180o • Facebook uses CNN for automatic tagging algorithms, • Amazon  for generating product recommendations and • Google  for search through among users’ photos Test Case #2 – MR1 Test Case #3 – MR2
  • 14. #ATAGTR2019 As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial channels(Provided due credit is given to me/us) 14 Use Case & Demo – Image Based Search in Social Media / Retail 34% 85% Consumers will spend more money online when AI is deployed effectively Customer interactions in retail will be managed by artificial intelligence, by 2020 Pinterest LensHow it should ideally work ? A Real life scenario
  • 15. #ATAGTR2019 As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial channels(Provided due credit is given to me/us) 15 Use Case & Demo pickle Serialize the data Trained Model Train dataset Test Dataset for prediction Shoe Images 60K predict ML Application CNN - ResNet TESTING FOR MACHINE LEARNING BASED IMAGE CLASSIFIER FOR SHOE BRAND RECOGNITION USING CNN What are the metamorphic relations in this use case ? Train & Test whether the CNN algorithm is able to detect the image as a shoe and the brand • Test Case 1: Happy Path Testing – Subset of the images used for training the algorithm • Test Case 2: Metamorphic Relation 1 - Input image as a rotated image of the shoe • Test Case 3: Metamorphic Relation 2 - Input image with BGR (instead of RGB) • Test Case 4: Testing with the data outside of the trained data set
  • 16. #ATAGTR2019 As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial channels(Provided due credit is given to me/us) 16 Use Case & Demo
  • 17. #ATAGTR2019 As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial channels(Provided due credit is given to me/us) 17 2. Testing For Metamorphic Relations – Testing for Unstructured Data Continuous Information flow Big Data Software Actionable Intelligence Sentiment Analysis L1 - Supervised L2 - Semi L3 - Unsupervised Use Case: Twitter Sentiment Analysis ML software Vocabulary Synonyms Antonyms Negations Test case #1 – MR1: Classification using should result Test case #2 – MR2: Classification using should result Test case #3 – MR3: Classification using should result Test case #4 – MR4: Classification using should result Testing entails validation of continuous information flow
  • 18. #ATAGTR2019 As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial channels(Provided due credit is given to me/us) 18 3. Adversarial Scenario Testing • Two-sample statistical hypothesis testing are conducted to determine both originate from the same distribution • Where there is high variation in distribution concludes influence of Adversial Inputs Adversarial examples are inputs to machine learning models that an attacker has intentionally designed to cause the model to make a mistake Adversarial Example: Adversarial Input + ML Model = High accuracy (>80%) or 150% more than normal ML model predictions Statistical Hypothesis Testing Attackers could target autonomous vehicles by using stickers or paint to create an adversarial stop sign that the vehicle would interpret as a ‘yield’ or other sign USE CASE
  • 19. #ATAGTR2019 As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial channels(Provided due credit is given to me/us) 19 4. Dual Coding Testing for Machine Learning Applications & Models DCT to validate Correctness of predictions by ML applications & models Dataset ML Model 1 ML Model 2 Trained with same or common set of features Compare predictions for correctness Prediction 1 using RF Prediction 2 using SVM Random Forest Support Vector Machine USE CASELEARNING MANAGEMENT SYSTEM - VDI MACHINE PREDICTION AI enabled LMS which has to predict & allocate VDI machines based on various factors including Course,# of Course Completed, # of pending course, Machine required. Test Data is passed through the 2 ML models namely Support Vector Machine & Random Forest. Random Forest consistently reported higher levels of accuracy and was chosen as the base model for implementation of LMS.
  • 20. #ATAGTR2019 As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial channels(Provided due credit is given to me/us) 20 Best Practices for Testing ML Applications ML TESTING BEST PRACTICES Acceptance criteria, with amount of error stakeholders are willing to accept Determine the level of outcomes acceptable for each scenario Communicate the level of confidence in the results Test with new data post training Scenario classification - Expected best case, Average case, worst case Defects will be reflected in the inability of the model to achieve the goals
  • 21. #ATAGTR2019 As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial channels(Provided due credit is given to me/us) 21 Goodbye GUI, Hello VUI • Voice First Device - Smart device designed to get tasks done conversationally • Always-ON and Always-listening - VPA enabled Wireless Speakers Analyst Corner (Gartner): • By 2021, early adopter brands that redesign their websites to support visual and voice search will increase digital commerce revenue by 30% • By 2019, half of major commerce companies and retailers with online stores will have redesigned their commerce sites to accommodate voice searches and voice navigation Voice Applications What are the Challenges in testing Voice applications? - Graphical User Interface (GUI) - Voice User Interface (VUI)  Prescriptive vs Descriptive  Limited data input vs indefinite # of test cases  Testing for languages vs Testing for languages & accents
  • 22. #ATAGTR2019 As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial channels(Provided due credit is given to me/us) 22 Testing the Voice User Interface Alexa Skill Test FrameworkTools & Frameworks Service Simulator Echosim.io Skill metrics “Alexa, Open Domino’s?” “. . . ” User Alexa enabled Devices Amazon – Echo, Dot, Tap, Fire TV Triby, Pebble Watch Alexa Voice Service (AVS) Alexa Skills Kit (Custom) Services Request Voice – device sends to AVS Response Voice Request HTTPS/JSON – POST request to Skill End point Response HTTPS/JSON Speech-to-Text technology AUT Alexa Skills Kit - create commerce driven skill Use Case – Leading Pizza retail chain using AI Voice Application Avoid giving word clues, Avoid think- aloud techniques, Avoid accidental wake-ups Certifies application functionality as per design specifications in all possible pathway and prompts Holistic, experimental review of a speech application; Test fully developed application through pre-determined use- cases Alexa Skills Kit (ASK) - Collection of self-service APIs, tools & more. Enables designers, developers & brands to build engaging skills & reach customers through tens of millions of Alexa Enabled Devices (AED). Dialog Traversal Testing (DTT) VUI Review Testing (VRT) Usability Inspection Testing
  • 23. #ATAGTR2019 As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial channels(Provided due credit is given to me/us) 23 Testing the Voice User Interface – Automated Voice Testing (AVT) Test Case Start Next End Text to Alexa Text-to-speech Speech-to-text Text-to-speech Text from Alexa Text to Alexa “Alexa, open Domino’s” “Welcome to Domino’s, what would you like to order?” “. . .” Speech Recognition Engine Echo  Codeless Automation tool - TestArchitect – Extend to C#, Java or Python  Based on Speech Recognition Engine (SRE)  Action-based, Interaction data table • Device Agnostic and platform agnostic • AVT can be applied to Alexa apps, Google Home apps, voice enabled web sites Features Benefits Sample Utterance • Variations of the sample utterances with different slot values and slightly different phrases • Slots for user errors: Test different permutations, such as missing or incorrect values • Provide multi accent intent, verify its capabilities to process the required ‘Intents’ & respond with the appropriate ‘Utterances’ Wake Word Alexa, Intent Slot To Plan my trip, For next Friday Test Scenarios Invocation name Ask Yatra,
  • 24. #ATAGTR2019 As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial channels(Provided due credit is given to me/us) 24 Automated Voice Testing (AVT) Demo
  • 25. #ATAGTR2019 As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial channels(Provided due credit is given to me/us) 25 Testing Strategy for Other Areas - Chatbots, AR/VR
  • 26. #ATAGTR2019 As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial channels(Provided due credit is given to me/us) 26 NLP : Chatbot Overview & Challenges A computer program that simulates human conversation through voice commands or text chats or both. What are the Challenges in testing a Chatbot?  Unstructured Input - There are no barriers for users in terms of asking questions  Multitude of user interactions - 100% of Test coverage is not possible  Non-deterministic Behavior - behavior keeps changing based on learnings What is a Chatbot ?
  • 27. #ATAGTR2019 As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial channels(Provided due credit is given to me/us) 27 Chatbot Testing • Conversational flow - Chatbots are based on conversations & very important to test conversational flow of a Chabot • Understanding - Process any kind of messages • Error Management – how does a chatbot react to errors • Conversational steps - minimize the number of steps in conversation • Bot speed - Speed at which the bot replies to the messages. • Bot Accuracy - Out of predicted utterances, # of utterances the bot gets correct will be accuracy of bot. Tools & Frameworks Functional Non Functional USE CASES Travel & Hospitality Bots can help customers plan and book trips, push personalized offers based on browsing history and preferences Content Distribution and News Push personalized news content, manage polls, unleash predictive content delivery based on behavior trends Health Care Patients leverage bots for appointments , receive personal updates based on treatment history etc.
  • 28. #ATAGTR2019 As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial channels(Provided due credit is given to me/us) 28 AR/VR Applications and Testing Challenges • Augmented Reality - Augments virtual elements into the real world • Virtual Reality – Digital creation of virtual environment around the user • AR/VR has a market potential of $95B by 2025 AR / VR Applications What are the Challenges in testing AR/VR applications?  Wide varieties of device conditions to be dealt with and Multiple compatibility Issues  High performing system – Performance challenges to create virtual / augmented word  Need for consistently rich user experience  Test Automation of AR/VR systems Magazine ads instantly shoppable - consumers purchase straight off the page Project their selected item (through video) in their living room to check suitability Alibaba offers VR shopping experience (BUY+) through mobile & VR Headsets - users get the experience of shopping in an actual store
  • 29. #ATAGTR2019 As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial channels(Provided due credit is given to me/us) 29 Testing Strategy for AR/VR Applications FUNCTIONAL TESTING USABILITY TESTING PERFORMANCE TESTING SYSTEM INTERACTION ASSURANCE • Validate Navigation, interactive buttons • Validate interaction with voice commands etc. AUGMENTATION ASSURANCE • Validate/Captures/ registers accurate dimensions of a 2D/3D object • Real time tracking NETWORK ASSURANCE • Test for varied network related issues (Wi- Fi/2G/3G/4G/5G ) GEO LOCATION – LOCAL AND GLOBAL • Ensure obtains local and global coordinates • Marker less apps integration with SLAM (simultaneous localization and mapping) COMPATIBILITY TESTING • For integrated and/or hosting devices and bandwidth EFFECTIVENESS • Match between system and the real world • Visibility of system status (real time) • User control and freedom EFFICIENCY • Consistency and standards • Error prevention • Accuracy • Recognition rather than recall (learnability) , Help users recognize, diagnose, and recover from errors • Environment configuration SATISFACTION • Aesthetic and minimalist design • Help and documentation • Customer experience with the system LATENCY TESTING • Test the quality of CGI and framerate for the visuals on stress conditions SCALABILITY TESTING • For cloud based AR/VR systems - dynamically scale the application resource to support increased demand PERFORMANCE UNDER DIFFERENT INPUTS • Assure performance for change in angle , orientation, distance between AR image and smart device/smart phone etc.
  • 30. #ATAGTR2019 As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial channels(Provided due credit is given to me/us) 30 AR/VR Automation AR/VR automation can be possible by building tools with latest technologies Computer Vision Robotics Device technologies will be integrated to Automation tools automate AR/VR Systems that functions based motion detection, to simulate user interactions AI based Computer vision libraries are going to help Automation Tools/Bots understand the visuals / response from AR/VR systems to automate the interactions. Robotic Motion Simulator Natural Language processing will be another Key element for the automation tools/bot to handle; AR/VR systems that accepts voice commands or allow voice interactions Natural Language Processing AI based bots that can monitor and learn human interactions with systems captured from Functional / Crowd / Beta testing and replicate for regression automation. Following are the key technologies that will allow the Tool to achieve AR/VR automation
  • 31. #ATAGTR2019 As a author of this presentation I/we own the copyright and confirm the originality of the content. I/we allow Agile testing alliance to use the content for social media marketing, publishing it on ATA Blog or ATA social medial channels(Provided due credit is given to me/us) 31 In Summary 1 2 3 4 1 . AI I S N E AR I N G M AI N S T R E A M AD O P T I O N AI has already picked up pace across various industry sectors with the aim of deriving business value in terms of enhanced customer experience, new sources of revenue and cost reduction 2 . T E S T I N G O F AI AP P S I S A C H AL L E N G E Non-deterministic and probabilistic nature, ever-changing behavior based on learnings and non-linear inputs (e.g. voice, conversational text, images/videos etc.). 3 . R O B U S T S T R AT E G Y TO AS S U R E AI Machine Learning Models (Feature Testing, Metamorphic Testing, Adversarial Scenario testing, Dual Coded ML etc.), Voice User Interfaces Testing and Chatbots Testing 4 . M E AS U R E T H R O U G H L E V E L O F C O N F I D E N C E Achieve the defined acceptance criteria and confidence levels to make the application available for wide scale adoption