Introduction

to AI
Excite Module
 What do you think Artificial Intelligence is?
 What do you want to learn about AI?
 How do you think we should go about it?
 What will you learn?
Artificial Intelligence
 It is concerned with the design of
intelligence in an artificial device
 It is the ability of a machine or a
computer program to carry out tasks that
normally requires human intelligence
Game reflection
 Semantris Quick Draw
Word association game Draw 6 objects correctly in a Rock Paper Scissors
powered by Machine row. Be the first to win 20 games
Learning. against the AI to obtain the
final clue.

Domain:
Domain: Domain:
Natural Language Processing
Computer Vision Data for AI
 Quiz Time
1. Which one of the 2. This language is easy to
following is an application learn and is one of the
of AI? most popular language
a. Remote controlled Drone for AI today:
b. Self-Driving Car a. C++
c. Self-Service Kiosk b. Python
d. Self-Watering Plant c. Ruby
System
d. Java
3. Which of the following is 5. What does NLP stand for
not a stage in the AI in AI?
Project Cycle: a. Neutral Learning
a. Problem Scoping Projection
b. Neuro-Linguistic
b. Data Acquisition Programming
c. Data Exploration c. Natural Language
d. Prototyping Processing
d. Neural Logic Presentation

4. This field is enabling 6. This is a program that

computers to identify and allows the computer to
process images like simulate conversation with
humans do: a human
a. Face Recognition being:
a. Speech Application
b. Model-view-controller Program Interface
c. Computer Vision b. Chatbot
d. Eye-in-Hand System c. Voice Recognition
d. Speech Recognition
7. This is a system of 9. What does model. Add
Programs and Data- (dense(32,
Structures that mimics the
input_shape=(784))) do?
operation of the
a. It adds an input layer
human brain:
b. It adds a hidden layer
a. Intelligent Network
c. It adds an output layer
b. Decision Support System
d. It adds a dense layer
c. Neural Network
d. Genetic Programming
10. How excited are you
8. Where is Decision tree about this AI curriculum?
used?
a. Very Excited!
a. Classification Problem b. A bit excited
b. Regression Problem c. Same as always
c. Clustering Problem d. Not excited at all
d. Dimensionality Reduction
Write a letter to your
future self
Right now, I am looking forward to…
Relate Module
Smart City
https://www.youtube.com/watch?v=eRMiK
t81nAE

Smart Home
https://www.youtube.com/watch?v=1Caja
UoI3vU
Activity
 Build your dream home
Activity
Write an interactive story
Using inklewriter

 https://www.inklestudios.com/inklewriter/
Purpose module
 how you can use AI for good-SDG
Possibilities module
Topics to be covered
 Understand the definition of AI
 Types of tasks done by AI
 Different components involved with intelligent
behaviour
 Examine the different ways of approaching AI
 Look at some example systems that use AI
 Trace briefly the history of AI
 What is AI
 It is concerned with the design of intelligence
in an artificial device
 It is the ability of a machine or a computer
program to carry out tasks that normally
requires human intelligence
 What is AI
 The Turing test was developed by Alan Turing
(Computer scientist) in 1950.
 Father of Artificial Intelligence: John McCarthy
 This term was coined by John McCarthy in 1956
Turing Test
 Types of Tasks done by AI
 Intelligententities needs to be able to do both
ordinary and expert tasks .
 Ordinary tasks:
 Planning activity,
 Recognizing people, objects
 Communicating
 Navigating round obstacles on the street
 Expert task:
 Medical diagnosis
 Mathematical problem solving
Expert task are comparatively easy for
machine to do than mundane task, as per
previous experiences in the field of AI
Intelligent behaviour comprise of

 Perception
 Reasoning
 Learning
 Understanding language
 Solving problems
Introduction to AI Domains
• Artificial Intelligence becomes intelligent according
to the training which it gets. For training, the machine
is fed with datasets.
• According to the applications for which the AI
algorithm is being developed, the data which is fed
into it changes.
• With respect to the type of data fed in the AI model,
AI models can be broadly categorised into three
domains:
Data Sciences
 Data sciences is a domain of AI related to
data systems and processes, in which the
system collects numerous data, maintains
data sets and derives meaning/sense out
of them.
 The information extracted through data
science can be used to make a decision
about it.
Computer Vision
 Computer Vision(CV) is a domain of AI that
depicts the capability of a machine to get and
analyse visual information and afterwards predict
some decisions about it.
 The entire process involves image acquiring,
screening, analysing, identifying and extracting
information.
 In computer vision, input to machines can be
photographs, videos and pictures from thermal or
infrared sensors, indicators and different sources.
 The main objective of this domain of AI is to teach
machines to collect information from pixels.
Natural Language Processing
 Natural Language Processing(NLP), is a branch of
artificial intelligence that deals with the interaction
between computers and humans using the natural
language.

 Natural language refers to language that is

spoken and written words by people
Homework
Que1. Categorize the following applications
into Data , CV or NLP

a) Face Lock in Smartphones-CV

b) Email filters-NLP
c) Amazon’s Alexa-NLP
d) PriceRunner -Data
e) Shopzilla-Data
f) FaceApp-CV
Question 2:
AI Ethics
 Moral Issues-Self driving Car
 Data Privacy
 AI Bias
 AI Access
Moral Issues-Self driving Car
Scenario :
 Let us now assume that the car has hit the boy
who came in front of it. Considering this as an
accident, who should be held responsible for it?
Why?

1. The person who bought this car

2. The Manufacturing Company
3. The developer who developed the car’s
algorithm
4. The boy who came in front of the car and got
severely injured
Data Privacy
AI Bias
 Another aspect to AI Ethics is bias.
 Everyone has a bias of their own no matter
how much one tries to be unbiased, we in
some way or the other have our own biases
even towards smaller things.
 Biases are not negative all the time.
 Any bias can transfer from the developer to
the machine while the algorithm is being
developed.
Some examples of AI bias:
 Majorly, all the virtual assistants have a
female voice.
 It is only now that some companies have
understood this bias and have started
giving options for male voices but since
the virtual assistants came into practice,
female voices are always preferred for
them over any other voice. Can you think
of some reasons for this?
 Ifyou search on Google for salons, the first
few searches are mostly for female salons.
 This is based on the assumption that if a
person is searching for a salon, in all
probability it would be a female.
 Do you think this is a bias? If yes, then is it
a Negative bias or Positive one?
AI Access
 AI creates unemployment
 AI for kids
AI Project Life Cycle
Problem Scoping

Data Acquisition

Data Exploration

Modeling

Evaluation
Problem Scoping

List of
Theme Topic
Problems
4 Ws Problem Canvas

Who What

Where Why
Problem Statement Template
Our [Stakeholder(s)] Who

Has/Have a [issues, problem, need] What

problem that

When/While [context, situation] Where

An ideal [benefit of solution for them] Why

solution would
Example problem for 4W canvas:

Discipline is an integral part of a schooling system. One of the main

objectives of sending a child to school is to make him/her
disciplined. Schools have their own rules and regulations which the
students need to follow in order to maintain the decorum of their
school. One of these regulations is around the uniform which
students wear to school. A lot of schools have incentivizing schemes
for students who wear proper uniform throughout the year so that
they can encourage everyone for the same. Still, it is observed that
many students do not wear proper uniform every day. Since the
strength of students in a school is way more that the strength of
teachers there, many times students escape out of any actions easily
and hence are not compelled to get disciplined.
Well, this is a problem. How can we encourage students to wear
proper uniform in school?
After reading this case study, draft the 4Ws problem canvas.
Data Acquisition
 Data can be a piece of information
or facts and statistics collected
together for reference or analysis.

 Whenever we want an AI project to

be able to predict an output, we
need to train it first using data.
Data Acquisition

 Data is of two types:

o Training Data
o Testing Data

 For better efficiency of an AI

project Training data should be
Relevant and Authentic
Difference between training and testing
data

Training Data Testing Data

 This is the one on  Test data is used
which we train and only to assess the
fit our model performance of
model.
basically to fit the
 Testing data is
parameters.
unseen data from
 Training data’s which the
output is available predictions have to
to model. be made.
 Data Features
 Data features refer to the type of data
you want to collect.
 This data can be text, images or real
time data .
Ways to acquire data
Some examples of authentic
sources Important Point

 data.gov.in Data
 india.gov.in which we
 Incometaxindia.gov.in collect should
 Indianrail.gov.in be
open-sourced
and not
someone’s
property
System Map
 System Maps are used to show the
components and boundaries of a system
at a specific point of time.
 Here, System is the goal of our project
and factor are the data features.

 https://ncase.me/loopy/
System Map to show Depression and
Anxiety Connection
Example of System Map to show Water Cycle
 Activity:
let’s think the type of data required for
the given AI model:

AI based system which gives access to the machine based

on authorized fingerprints

AI based system which unlocks the door by recognizing

your face.

AI based system which predicts the temperature in the

month of December this year
Data Features
Let’s list down the data features for the given
system/problem scoped:

How driving crowds affects walking and cycling?

Photosynthesis
System map
 QA
1. What is a problem statement template
and what is its significance?

Ans: The problem statement template gives

the clear idea about the basic framework
required to achieve the goal. It is the 4Ws
canvas which segregates; what is the
problem, where does it arise, who is
affected, why is it a problem? It takes us
straight to the goal.
2. Mention the precautions to be taken
while acquiring data for developing an
AI Project.
3. What do you mean by Data Features?
4. Write the data features for the following:
Photosynthesis
5. Write the names for missing stages in the
given AI project cycle:

6. Do ethics in AI hamper data acquisition

stage? Justify your answer.
7. Why are system maps used in data
acquisition?
8. Name any two open source government
portals
Data Exploration
Data Exploration
To analyse the data, we need to visualise it
in some user-friendly format so that we can:

 Quickly get a sense of the trends,

relationships and patterns contained
within the data.
 Define strategy for which model to use at
a later stage.
 Communicate the same to others
effectively.
To visualise data, we can use various types of
visual representations.

 https://datavizcatalogue.com/
 Types
Line Graphs are used to display quantitative values
over a continuous interval or time period.
Example:
The following line graph shows annual sales of a
particular business company for the period of six
consecutive years:
Types
Bar Charts represent categorical data with
rectangular bars. They are commonly used to
compare several categories of data.
Example:
The bar chart below represents the total sum of
sales for Product A and Product B over three years.
 Types
Pie Charts It shows part-whole relationships. It
displays data and statistics in an easy-to-understand
‘pie-slice’ format and illustrates numerical
proportion.

Example:
The pie chart below
represents the proportion
of types of transportation
used by 1000 students to
go to their school.
QA
 Name the third stage of AI project life
cycle.
 Why data analysis is required?
 Enlist any 6 types of graphs?
 Why graphs considered better to analyze
data rather than text data?
QAs
1. The basis of decision making depends upon the
availability of ____________and how we experience and
understand it.
2. A machine can also become intelligent if they are
trained with __________ which helps them achieve their
tasks
3. What do you understand by linguistic Intelligence?
4. Mention two types of machines which have evolved with
time.
5. Mention four examples of artificially intelligent
applications in our smartphones.
6. How does a machine become Artificially Intelligent?
7. Mention four examples of machines that are not AI but
confused with AI.
8. Pick the odd one out and justify your answer:
 a. Snap Chat Filter b. Face Lock in Phone
 c. Chatbot d. Image search Option

9. Explain how AI works in the following areas:

a. Google Search Engine
b. b. Voice Assistants
c. c. E-commerce websites

10. Why training with information/Data is important in Artificial Intelligent

devices?

11. Differentiate between what is AI and what is not AI with the help of
an example?

12. How AI helps in giving you personalized experience online?

Revision
1. Give any two example applications of Data science?
2. Name any two examples of Computer vision?
3. Name any two examples of Natural Language
Processing?
4. Where do we collect data from?
5. What do you understand by AI bias?
6. Should AI replace laborious jobs? Is there an alternative
for major unemployment?
7. List four AI Ethics.
8. What are sustainable development goals?
9. Name the 4Ws of problem canvases under the problem
scoping stage of the AI Project Cycle.
10. What is the objective of evaluation stage?
11. Name all the stages of an AI Project cycle.
Modeling
 It refers to developing algorithms, also
called models which can be trained to
get intelligent output. That is, writing code
to make a machine artificially intelligent.
Generally, AI models can be classified
as follows:
Rule Based-Decision Tree
 It
is a rule-based AI model which
helps the machine in predicting
what an element is with the help of
various decisions (or rules) fed to it.
Examples of rule based models
 Decision Tree
 expert system help a doctor to choose
the correct diagnosis based on symptoms
 Rule based games to take moves
Rule Based-Decision Tree
 Points to Remember
 While making Decision Trees, one should take a good
look at the dataset given to them and try to figure out
what pattern does the output leaf follow. Try selecting
any one output and on its basis, find out the common
links which all the similar outputs have.
 Many times, the dataset might contain redundant data
which does not hold any value while creating a decision
tree. Hence, it is necessary that you note down which are
the parameters that affect the output directly and should
use only those while creating a decision tree.
 There might be multiple decision trees which lead to
correct prediction for a single dataset. The one which is
the simplest should be chosen as the best.
Create a decision tree by observing the
data given below in the table:

Age Adventurous Car Type

>=35 Yes Hummer Car

<35 No Civic
>=35 No SUV
<35 YES Sports Car
AI, ML, DL
 Artificial
Intelligence
 Machine Learning
 Deep Learning
AI, ML, DL
 Artificial Intelligence
 It refers to any technique that enables
computers to mimic human intelligence.
 It gives the ability to machines to recognize a
human’s face; to understand the voice
commands by humans, and also do other
tasks.
 The AI-enabled machines think
algorithmically and execute what they have
been asked for intelligently.
Machine Learning (ML)

 Machine Learning, or ML, is a subset of

Artificial Intelligence ,enables machines to
improve at tasks with experience. The
machine learns from its mistakes and
takes them into consideration in the next
execution. It improvises itself using its own
experiences.
 The intention of Machine Learning is to
enable machines to learn by themselves
using the provided data and make
accurate Predictions/Decisions.
Deep Learning (DL)
 Itenables software to train itself to
perform tasks with vast amounts of data.
 In Deep Learning, the machine is trained
with huge amounts of data which helps it
in training itself around the data.
 Such machines are intelligent enough to
develop algorithms for themselves.
a) Define Modelling. Explain the relationship between AI, ML
and DL.
b) Differentiate between Learning based and Rule based
approach.
c) What is the drawback of rule based approach?
d) What makes a machine work intelligently?
e) Why should the training data set be relevant and accurate?
f) What is the need of data exploration? Enlist any 6 methods
of data visualization.
g) How AI helps in giving you personalized experience
online?
h) What is Testing Dataset?
i) Mention the types of learning approaches for AI modelling.
j) What is a problem statement template and what is its
significance?
 To understand how the machines learn, let us
first look at three common ways that humans
learn.

Supervised Learning

Reinforcement Learning

Unsupervised Learning
Supervised Learning
Supervised Learning
 Insupervised learning, an algorithm is
employed to learn the mapping function from
the input variable (x) to the output variable (y);
that is y = f(X).
 Two types: Classification and Regression
Supervised Learning
 Similarity:
 Regression and classification are categorized
under the same umbrella of supervised machine
learning.
 Both share the same concept of utilizing known
datasets (referred to as training datasets) to make
predictions.

 Difference
 The main difference between them is that the
output variable in regression is numerical (or
continuous) while that for classification is
categorical (or discrete).
Supervised Learning
 Example:
 when provided with a dataset about houses,
and you are asked to predict their prices, that
is a regression task because price will be a
continuous output.
 when provided with a dataset about houses,
a classification algorithm can try to predict
whether the prices for the houses “sell more
or less than the recommended retail price.”
Supervised Learning
Google: train your model
To do activity

 https://teachablemachine.withgoogle.co
m/train
Identify the following as classification
and regression learning

 Differentiates between an orange and an

apple
 To predict your next salary, based on your
previous salary, any increments
 To check an email is spam or not spam
Unsupervised Learning
 Unsupervised Learning
 Unsupervised learning is where you only have
input data (X) and no corresponding output
variables.
 The unsupervised learning models are used to
identify relationships, patterns and trends out of
the data which is fed into it.
 These are called unsupervised learning because
unlike supervised learning, there is no correct
answers and there is no teacher. Algorithms are
left to their own plans to discover and present the
interesting structure in the data.
 Unsupervised Learning
 Two types of it: Clustering & Dimensionality
Reduction
 Clustering deals with finding a structure in a
collection of unlabeled data.
 Clustering can be defined as “the process of
organizing objects into groups whose members
are similar in some way”.
 A cluster is therefore a collection of objects which
are “similar” between them and are “dissimilar” to
the objects belonging to other clusters.
Unsupervised Learning
Unsupervised Learning
 Dimensionality Reduction is a technique in
Machine Learning that reduces the
number of features in your dataset.
 dimension reduction does not negatively
affect your machine learning model's
performance.
 In some cases, this technique has even
increased the accuracy of the model.
Unsupervised Learning
 Infinite drum machine experiment

 https://experiments.withgoogle.com/dru
m-machine
Neural Network Activity

 A neural network is a network or circuit of neurons,

or in a modern sense, an artificial neural network,
composed of artificial neurons or nodes.
 A neural network is essentially a system of
organizing machine learning algorithms to
perform certain tasks.
 The key advantage of neural networks, are that
they are able to extract data features
automatically without needing the input of the
programmer.
 It is a fast and efficient way to solve problems for
which the dataset is very large, such as in images.
Larger Neural Networks tend to perform better with larger
amounts of data whereas the traditional machine learning
algorithms stop improving after a certain saturation point.
Biological Neural Network
Similarities based on the terminology
between BNN and ANN
Application
 Neural networks are used for solving many
business problems such as
 sales forecasting,
 customer research,
 data validation, and
 risk management
 Speech recognition
 Character Recognition
 Signature verification
 Human face recognition
For the above general model of artificial neural network, the net
input can be calculated as follows −
yin=x1.w1+x2.w2+x3.w3…xm.wm
i.e., Net input yin=∑mixi.wi
The output can be calculated by applying the activation function
over the net input.
Y=F(yin)
Output = function netinputcalculated
 Questions
1. Differentiate between Rule based and Learning
based AI modeling approach.
2. Write one disadvantage of Rule based
approach.
3. Mention the types of learning approaches for AI
modeling.
4. Give any two examples of Data science,
Computer vision, Natural Language Processing?
5. What do you mean by machine learning?
6. Name any two examples of Machine Learning?
7. Examples of rule based approach.
8. Snapchat filters use _____ and _____ to enhance your
selfies with flowers, cat ears etc.
a. machine learning and deep learning
b. data and image processing
c. augmented reality and machine learning
d. NLP and computer vision

9. Based on the image below, choose the correct domain or

domains of AI required for it:

a) Data
b) NLP
c) Computer Vision
d) Both (a) and (b)
10. Rock paper and scissors game is based on the
following domain:
a. Data for AI
b. Natural Language Processing
c. Computer Vision
d. Image processing
11. Select a game which is based on Data
Science domain of AI:
a. Rock Paper and Scissors
b. Mystery Animal
c. Emoji Scavenger Hunt
d. Pokémon
13. Identify the domain of AI in the following image:

a. Data for AI
b. Natural Language Processing
c. Computer Vision
d. Rule based
14. Where do we collect data from?
15. Why do we need to collect data?
16. What do you understand by Data Privacy?
17. Is data which is collected by various applications
ethical in nature? Justify your answer.
18. What do you understand by AI, ML & DL? How
are they different from each other?
19. Why do apps collect data in our phone?