Emerging technologies - Digital Transformation

Digital Transformation Technologies

1. UNIT II
Digital Transformation Technologies

2. Topics
Review of Artificial Intelligence
Review of Machine Learning
Review of Artificial Neural Networks
Democratized AI
Cognitive Computing
Affective Computing and Emotional AI
Understanding Computer Vision
Conversational Platforms
Basics of Virtual Reality, Mixed Reality, Immersive
Reality
Experiencing Augmented Reality

3. Introduction
IMAGINATION
REALITY

4. Introduction

5. Introduction
Source Image from Oracle Blog
https://blogs.oracle.com/bigdata/difference-ai-machine-learning

6. Review of Artificial Intelligence
Whatis?
ArtificialIntelligence(AI)isthereplicationofhuman
analyticaland/ordecision-makingcapabilities.
Theterm"ArtificialIntelligence"referstothe
simulationofhumanintelligenceprocessesby
machines,especiallycomputersystems.Italso
includesExpertsystems,voicerecognition,
machinevision,andnaturallanguageprocessing
(NLP).

7. What is Artificial Intelligence ?
making computers that think?
the automation of activities we associate with human
thinking, like decision making, learning ... ?
the art of creating machines that perform functions that
require intelligence when performed by people ?
the study of mental faculties through the use of
computational models ?

8. Review of AI
The truth is, there are multiple ways for the
different types of AI to be classified
Basic Classification :
Technology-based
Functionality-based
Capability-based

9. Technology based
The technology-based categorization is likely one of
the most well-known ones in business circles

10. What ? How? Where?
RPA :
What?
To automate repetitive tasks and carry them out without human
intervention with the help of artificial intelligence or metaphorical
software robots
How?
First identifying repetitive tasks suitable for automation, then designing,
testing, and deploying bots to perform these tasks using specialized
RPA software.
Business Cases : Invoice Processing, Customer Service, Order
Fulfillment, Compliance Reporting
Tools : UiPath, Automation Anywhere, Blue Prism, Kofax RPA

11. What ? How? Where?
Speech Recognition
What?
Speech recognition programs are those that process human speech into
textual format an these are not voice recognition systems.
How?
Speech recognition is implemented by using algorithms to convert spoken
language into text, involving processes such as acoustic modeling, language
modeling, and signal processing.
Business Use Cases of Speech Recognition : Customer Service, Healthcare
Documentation, Virtual Assistants
Tools : Google Cloud Speech-to-Text, Amazon Transcribe, IBM Watson Speech
to Text, Nuance Dragon

12. What ? How? Where?
Computer Vision
Computer vision systems leverage artificial intelligence to process digital
images, videos, and other visual inputsto extract meaningful information that
can serve as a basis for decision-making.
How?
Computer vision is implemented by using machine learning and deep learning
algorithms
Business Use cases
Quality Inspection in Manufacturing, Retail Analytics, Medical Imaging,
Autonomous Vehicles
Tools : OpenCV, Tensotflow, Microsoft Azure Computer Vision, Amazon
Rekognition

13. What ? How? Where?
Generative AI
Generative artificial intelligence is a cutting-edge technology that harnesses
the power of machine learning to produce new content, mimicking and
sometimes surpassing human creativity.
How?
Generative AI is implemented by training models using large datasets to
learn patterns and generate new content, involving techniques such as
generative adversarial networks (GANs) and variationalautoencoders
(VAEs).
Business Use cases : Content Creation, Product Design, Drug Discovery,
Customer Personalization
Tools : OpenAI GPT, DeepArt, Runway ML, Runway ML

14. What ? How? Where?
Data Science
The termsdata scienceand artificial intelligence are often used
interchangeably to describe technologies that process data and extract insights
from it.
How?
Data science is implemented by collecting, cleaning, and analyzing large
datasets using statistical methods, machine learning algorithms, and data
visualization tools
Business Use cases
Predictive Analytics, Customer Segmentation, Fraud Detection, Supply Chain
Optimization
Tools : Python, R, Jupyter Notebook, Apache Spark, Hadoop, Mapreduce

15. Context
Context
A large retail chain aims to transform one of its
flagship stores into a smart store to enhance
customer experience, optimize operations, and
improve decision-making. This involves
integrating RPA, speech recognition, computer
vision, generative AI, and data science
technologies.

16. Context
Robotic Process Automation (RPA)
Task Automation: Implementing RPA to handle
repetitive tasks such as order processing,
inventory management, and supplier
communications. Bots will automate the entry of
purchase orders into the ERP system, reconcile
invoices, and manage stock replenishment.
Customer Service: Deploying RPA bots to
manage customer service requests, process
returns, and update customer records in real-
time.

17. Context
Speech Recognition
Customer Assistance: Integrating speech
recognition into in-store kiosks and mobile apps
to allow customers to ask for product
information, check stock availability, and receive
personalized recommendations through natural
language queries.
Employee Productivity: Equipping store
associates with speech-to-text tools for quick
note-taking and real-time transcription of
customer interactions, improving service
efficiency and accuracy.

18. Context
Quality Inspection: Using computer vision for
automated quality checks on products as they are
restocked, ensuring high-quality standards and
reducing manual inspection efforts.
Retail Analytics: Analyzing customer behavior
through video surveillance to optimize store
layouts, manage crowd control, and personalize
marketing displays based on foot traffic patterns.
Loss Prevention: Implementing real-time
surveillance systems to detect and prevent theft,
enhancing security and reducing inventory
shrinkage.

19. Context
Content Creation: Utilizing generative AI to
automatically create and update product
descriptions, marketing materials, and social
media posts, ensuring timely and engaging
content for customers.
Product Design: Leveraging generative AI to
simulate new product designs and displays,
enabling rapid prototyping and testing of store
layouts and product arrangements.
Customer Personalization: Creating
personalized shopping experiences by generating
tailored recommendations and offers based on
customer preferences and shopping history.

20. Context
Data Science
Predictive Analytics: Analyzing historical sales
data and customer behavior to forecast demand,
optimize inventory levels, and plan promotional
activities, ensuring products are available when
and where customers want them.
Customer Segmentation: Segmenting
customers based on purchasing behavior,
demographics, and preferences to deliver
targeted marketing campaigns and personalized
shopping experiences.
Fraud Detection: Applying machine learning
models to monitor transactions and detect
fraudulent activities in real-time, protecting the

21. Context Important

22. Functionality Based AI
This classification breaks down the different types of artificial intelligence software
based on the function that they perform but also places an emphasis on the machine’s
similarity to the human mind.

23. Reactive Machines
What?
Basic AI type, Does not have memory, They don’t learn from past
experiences or form memories
How?
Defining Rules: Establishing a set of rules and responses based
on possible inputs.
Programming: Coding these rules into the system, ensuring it can
recognize and react to the defined stimuli.
Testing: Rigorously testing the system to ensure it reacts
appropriately in all expected scenarios.
Business Cases : Manufacturing Process Control, Automated
Assembly Lines, Automated Response Systems, Non-Player
Characters (NPCs)
Tools : State Machines, Rule based (Drools or JBoss Rules), ROS
(Robot Operating System)

24. Limited Memory
What ?
Limited memory AI refers to models that have some memory
retention capabilities but to a limited extent
How?
Data Collection, Model Training, Real-Time Processing, Decision
Making
Business Use cases : Navigation and Control, Adaptive Cruise
Control, Chatbots with Context, Patient Monitoring, Predictive
Diagnostics, Dynamic Pricing, Inventory Management
Tools : TensorFlow, Keras, Scikit-learn

25. Theory of Mind
What? Theory of mind AI refers to a machine’s
ability to make decisions in a manner that humans
do –Reasoning
How ? NLP, Emotion Recognition, Cognitive
Modeling, Context Awareness
Business Use cases : Mental Health Support,
Human like interaction, Adaptive Learning,
Behavior Analysis
Tools : IBM Watson, Affectiva, Microsoft Azure
Cognitive Services, Replika(chatbot)

26. Self Awareness
What ? This means they are capable of
understanding their own states, emotions, and
existence
How? Self-Modeling: Introspection: Emotional
Understanding, Contextual Awareness
Business Use cases : Autonomous Surgical Robots, Predictive
Maintenance, Self-Driving Cars, Intelligent Virtual Assistants
Tools : TensorFlow, Keras

27. Capability Based AI

28. ANI
What? these devices perform very narrow,
specific tasks and are thus largely based on
reactive machines or limited memory AI models.
Voice Assistants : Siri, Alexa, and Google
Assistant
Recommendation Systems : Netflix and
Amazon Recommendations
Image Recognition : Facial Recognition
Systems

29. AGI
Artificial general intelligence is also sometimes
referred to as strong or deep AI, and it relates to
machines that are able to mimic human
intelligence and behavior.
Only in Pilot mode
AGI is based on the theory of mind framework
General-Purpose Robotics, Comprehensive Life
Management, Autonomous Research and
Development

30. Artificial Super Intelligence
Artificial Super intelligence (ASI) refers to a
hypothetical form of artificial intelligence that
surpasses the most intelligent human minds
across all domains
Global Problem Solving, Advanced Scientific
Research, Next-Generation Technologies,
Superintelligent Partners
Artificial Superintelligence remains purely
theoretical and is the subject of extensive debate
and speculation.

31. more powerful and more useful computers
new and improved interfaces
solving new problems
better handling of information
relieves information overload
conversion of information into knowledge
Some Advantages of Artificial
Intelligence

32. The Disadvantages
increased costs
difficulty with software development -slow and
expensive
few experienced programmers
few practical products have reached the market as
yet.

34. Machine Learning

37. Getting into
Machine Learning

38. Getting into
Machine Learning

39. Getting into
Machine Learning

40. Getting into
Machine Learning

41. Getting into
Machine Learning

42. Few Applications of
Supervised Learning

43. Getting into
Machine Learning

44. Getting into
Machine Learning

45. Few Applications of Un-
Supervised Learning

46. Getting into
Machine Learning

47. Getting into
Machine Learning

48. Getting into
Machine Learning

49. Few Applications of
Reinforcement Learning

50. Getting into
Machine Learning

51. Top
Applicati
ons of
ML
•Traffic Alerts
•Social Media
•Transportation
and Commuting
•Products
Recommendation
s
•Virtual Personal
Assistants
•Self Driving Cars
•Dynamic Pricing
•Google Translate
•Online Video
Streaming
•Fraud Detection

52. Top
Companies
uses ML
•IBM Watson
•Facebook
•Facebook
Chatbot
•Pinterest
•Twitter
•Google
•Amazon
•Youtube

53. Review of Artificial Neural Networks
https://www.youtube.com/watch?v=bfmFfD2RIcg
&t=33s
Refer Word Document –Self Learning

54. Democratized AI
https://www.youtube.com/watch?v=LZWl8RiTE2I