ARTIFICIAL INTELLIGENCE

•Artificial Intelligence (AI) refers to the

ability of machines to perform tasks like
thinking, perceiving(noticing), learning,
problem-solving, and decision making.
 ARTIFICIAL INTELLIGENCE

• Artificial means "man-made" and Intelligence

means "ability to think and take decisions".
• In simple words, Artificial refers to the things
made by humans and the Intelligence refers to
the ability of people to think and learn from their
past experience.
 A.I.
AROUND US
 SMARTPHONES
• The smartphone has many applications that running and
provided services with the help of AI.
Ex. Google Assistant,
Alexa, Apple Siri, etc.
 SOCIAL MEDIA
• Social media websites like Twitter,
Facebook, Instagram, or Snapchat
Sending notification and managing
timelines by AI.
• AI takes all your past behaviour, web
searches, interactions, and everything
else that you do when you are on these
websites and tailors the experience just
for you.
MUSIC AND MEDIA STREAMING

• Apps like Spotify, NetFlix, or

Youtube AI is making a
decision for the users.
• AI records playlist history
and generating some
recommendations for
watching or playing songs.
 VIDEO GAMES
• Video games companies are most
earlier adopters of AI. AI generate
random levels in video games.
• In many games, AI defeated world
champions. PUBG, Dota 2, Fortnite
all are AI integrated games.
 SMART HOME
• Many smart home devices use AI to
learn the behaviour of the members
of the family and can adjust settings
accordingly.
• Smart voice assistants playing a vital
role in smart homes.
• Smart thermostats used to adjust the
temperature based on the user's
preferences.
• Smart lights change the colour and
intensity of lights based on time and
much more.
 SECURITY AND SURVEILLANCE
• Thousands of cameras
keep monitoring at the
same time by AI only.
• Object recognition and
face recognition getting
better and better day by
day.
 Smart Keyboard and Apps
• Smart Keyboards provide comfort for users while typing
on the screen.
• It generates suggestions based on the writing style of
users.
• It also displays a few words and emojis.
 Healthcare

• With an introduction to AI-

powered machines detection of
disease and treatment becomes
a bit easier and convenient.
• AI-powered machines make the
process of treatment and
management simplified research
to cure some disease done by
AI- based systems.
 E-Commerce

• Online shopping on
Amazon and eBay like
websites using chatbots
to collect data of
customers and building a
good rapport with buyers.
 Smart Email

• Modern email apps like spark

provides the facility to get rid of
spam email and unwanted emails.
• It also categorizes email, so users
can quickly read the important
ones.
• The smart reply concept also
giving a few suggestions with a
reply text like in Gmail.
 SMART CARS

• Tesla is a prime
example of AI is
impacting in our
daily life.
 SMART DRONES
• Companies like Amazon and
Walmart are heavily
investing in drone delivery
programs and it will become
a reality far sooner than
what you expect.
 Banking and Finance
• The banking and finance industry
relies on AI for providing customer
services, protection against fraud,
investment suggestions, and so
on.
• While using the chat service of
banks the chat is represented by
Bots only. In the finance industry,
AI is used to analyze data.
 ONLINE ADS NETWORK
• AI just not tracking records of users but also serve the ads
based on statistics.
• With the help of AI Ads network displaying random Ads
online.
 Navigation and Travel

• While traveling or enjoying

rides like ola, uber, or any
other services, google map
navigation help to find a
perfect route for the journey.
• Moreover, AI can give you
real-time traffic data.
 DATA SCIENCE
If a person thinks of automate any system or want a report or analysis
of customers' feedback, data is required.
For example: Taking student's daily attendance we need data of
students like class, roll number, student name, etc.
Data can be the backbone of AI. Almost 98% of AI systems are
dependent on data. As system development grows processing of data
also increases. This data can be in any form textual information, audio,
video, big data like predictions, insights, forecasts, decision making,
etc.
 Computer Vision (CV)

It is a field of science that deals with how computers gain

a high level of understanding from digital images or
videos. It is a flown that studies how the human visual
system works. The computer vision includes the following
methods to produce information:
1. Acquiring Images 2. Processing Images
3. Analysing Images 4. Understanding Images
 Computer vision

Computer Vision is mainly used for Face recognition systems to recognize

the faces in images and videos. The application areas like google photos,
spam chat, Facebook, Instagram etc.
Content-Based Image Retrieval systems identify images based on image
properties like composition, colour, texture etc. The application areas are
search engines like google & bing, used in different CT scans and MRIs in
hospitals, etc.
Computer Vision also helpful in smart interactions to supply input to
computers. It is mainly used in games, systems designed for differently-
abled individuals, etc.
 Computer vision

Computer vision also helps in Environment

Perception such as analyzing videos, images, or
video feeds for identifying patterns and perceiving
the environment. Application areas are Home
security systems, Office security systems, Drone-
based surveillance systems etc.
 Natural Language Processing (NLP)
The programming languages work on their own principles, syntax, and
keywords. The aim of NLP is developing such systems that work on human
natural language on oral as well-spoken language.
It has two main components:
1. Natural Language Understanding (NLU): It is used for spoken or written
language to provide a link between natural language inputs and what they
present. It analyses different aspects of language.
2. Natural Language Generation (NLG): It helps to produce meaningful
phrases and sentences along with Text planning, Sentence Planning, and
Text realization.
AI project cycle is the process of converting
the real-life problem into an AI-based model.
The project cycle framework is designed to
help project managers guide their projects
successfully from start to end. The purpose
of the project cycle is to create an easy-to-
follow framework to guide projects. The AI
project cycle provides us with an appropriate
framework which can lead us towards our
goal.
STAGES OF AI PROJECT CYCLE
Problem Scoping is the first stage of the AI project
cycle. In this stage of AI development, problems will
be identified. It is then followed by designing,
developing, or building, and finally testing the project.
In AI project cycle everything will be failed if problem
scoping is failed or without appropriate problem
scoping. Incorrect problem scoping also leads to
failure of the project as well.
What is Problem Scoping?
Whenever we are starting any work, certain problems
always associated with the work or process. Actually we
are surrounded by problems! These problems can be small
or big, sometimes we ignore them, sometimes we need an
urgent solution otherwise your work will suffer.

The problem scoping refers to

the identification of a problem
and the vision to solve it.
Sustainable development means developing
cities, land, businesses, and communities to
meet the needs of the present, without
effecting future generations’ ability to meet
their needs.
The environment underpins each of the
SDG’s – they seek to improve living
conditions for all, without increasing the use of
natural resources.
The SDG’s work to protect the planet’s
resilience for our future generations.
 17 Sustainable Development Goals:

No poverty Reduced inequalities

Zero hunger Sustainable cities & communities

Good health & well-being Responsible consumption & production

Quality education Climate action

Gender equality Life below water

Clean water & sanitation Life on land

Affordable & clean energy Peace, justice & strong institutions

Decent work & economic growth Partnerships

Industry, innovation & infrastructure

The 4Ws of Problem Scoping
The 4Ws are very helpful in problem scoping. They
are:

Who? - Refers that who is

facing a problem and who are
the stakeholders of the problem
What? - Refers to what is the
problem and how you know
about the problem.
Where? - It is related to the
context or situation or location
of the problem
Why? - Refers to why we need to
solve the problem and what are the
benefits to the stakeholders after
solving the problem.
 Your final problem statement will
look likes the following table:
Stakeholders
Who Farmers, Fertilizer Producers, Labours, Tractor
Companies
The problem, Issue, Need
What Determine what will a good time for seeding or crop
harvesting?
Context/Situation
When Decide the mature age for the crop and determine its Time
Benefits
Ideal
Take the crop on time and supply against market demand on
Solution time
The problem statement template

When the above 4Ws are completely filled you

need to prepare a summary of these 4Ws. This
summary is known as the problem statement
template. This template explains all the key
points in a single template. So if the same
problem arises in the future this statement helps
to resolve it easily.
 Set up the goal
After understanding and writing the problems, set your goals, and
make them your AI project target. Write your goals for your selected
theme.
Suppose you have selected theme of agriculture then write how AI will
help farmers to solve their problems.
• Determine what will a good time for seeding?

• Determine what will be a good time for harvesting?

• Determine when and how much fertilizer will be applied to the

selected crop?
These goals can be more!

Now think and apply the 4Ws strategy for each problem or goal.
Data Acquisition consists of two words:

Data : Data refers to the raw facts , figures, or

piece of facts, or statistics collected for reference or
analysis.

Acquisition: Acquisition refers to acquiring data

for the project from reliable/relevant sources.
 TYPES OF DATA
• Training Data: The collected data through
the system is known as training data. In
other words the input given by the user in
the system can be considered as training
data.

• Testing Data: The result data set or

processed data is known as testing data. In
other words, the output of the data is known
as testing data.
 Reliable sources of relevant data:
SURVEYS Data can be collected from online surveys, telephonic surveys or in person surveys
and collect responses. Surveys are a way of collecting data from a group of people in order to
gain information and insights into various topics of interest. The process involves asking people
for information through questionnaires which can be online or offline. It can be considered as
a data source.
WEB SCRAPING: Data or information can also be extracted from a website. Web scraping or
Data scraping is the method of downloading information from the World Wide Web (WWW)
and storing it onto your computer for later reference. The data collected in this way is an
online data.
SENSORS: Data can also be collected from various sensors like collecting environmental data
and stored in some data storage solutions. Sensors are connected through gateways which
enable them to collect live data in the offline mode
CAMERAS: Data can be seen, written down or recorded onto the computer: Cameras are used
to collect data in the form of images. CCTV, web cameras, surveillance cameras are big sources
of visual data that can be acquired from various places.
 Reliable sources of relevant data:
OBSERVATIONS: It is a method of collecting data by watching facts as they occur.
Using the observation technique. data can be analysed and used for testing the
model.
APPLICATION PROGRAMMING INTERFACE (APIS): APIs are a set of functions and
procedures that allow one application to connect to another. So, one of the ways
of collecting data is through APIs that can be used to collect data from social
media services for analysis. There are times using the internet, we acquire
unauthentic data from websites for our Al project. Extracting private data can be
an offence. So, keeping this in mind we should ensure the data is collected from
open-sourced websites hosted by the government. They are one of the most
reliable and authentic sources of information These portals have information
collected in suitable format that can be downloaded. Some of the open-sourced
government portals are data.gov.in, india gov.in, etc.
DATA EXPLORATION/
DATA VISUALISATION
Data Exploration refers to the techniques and tools
used to visualize data through complex statistical
methods. It is the graphical representation of data
and information. By using graphical tools like charts
and graphs, it is an easier way to understand the
trends and patterns of data.
Advantages of Data Visualization
A better understanding of data
Allows user interaction.
Provide real-time analysis.
Help to make decisions.
Reduces complexity of data
Provides the relationships and patterns contained
within data
Define a strategy for your data model
Provides an effective way of communication among
users
Till now you learned about problem scoping and data acquisition.
Now you have set your goal for your AI project and found ways to
acquire data. When you acquired data the main problem with data
is - the data is very complex. Because it's having numbers. To make
use of these numbers user need a specific pattern to understand
the data.

For example if you are going to reading a book. You went to library
and selected a book. The first thing you try to do is, just turning
the pages and take a review and then select a book of your choice.
Similarly, when you are working with data or going to analyze data
you need to use data visualization.
Data Visualization Tools
There are many data visualization tools available.
Here I made a list of 20 data visualization tools for
you. Although there are many more tools available and
these numbers increasing day by day.
How to select a proper graph?
We should select an appropriate chart for data visualization.
The selection of chart all depends on the data and the goal
you are going to achieve through your model. Although some
basic purposes of charts that let you select an appropriate
chart, they are as follows:
Comparison of Values - Show periodical changes
i.e. Bar Chart
Comparison of Trends - Show changes over a period of time
i.e. Line Chart
Distribution of Data according to categories - Show data
according to category i.e. Histogram

Highlight a portion of a whole - Highlight data according to

value i.e. Pie Chart

Show the relationship between data - Multiple charts can

be used
 MODELLING
we can represent data in graphics using
various tools. This graphical representation
makes data easy to understand for the
humans to take a decision or prediction. But
when it comes to machine to access and
analyse data, machine requires mathematical
representation of data. Hence every model
needs a mathematical approach to analyse
data.
AI modelling approaches

Basically there are two approaches broadly taken by

researchers for AI modelling. They are:

Rule-Based Approach

Learning-Based Approach
 RULE BASED APPROACH

A Rule-based approach is
generally based on the data and
rules fed to the machine, where
the machine reacts accordingly
to deliver the desired output.
Rule-based learning follows the
relationship or patterns in data defined
by the developer.
The machine follows the instructions
or rules mentioned by the developer
and performs the tasks accordingly.
It uses coding to make a successful
mode.
 Suppose you have data of 100 employees and 100 businessmen. The
following steps you need to follow to train your machine:

1. Input your data and label them accordingly for employees and
businessman.
2. Now if the data is related to employee, the machine will compare its
rules defined by you as employee and label it as employee and this way
it will identify the data of employee.
3. Similarly it will follow the rules for businessman as well.

Here in the machine, you need to feed some of the characteristics of

employees like earning money and provide service whereas businessman
investing money and provide service to train the machine.
 Example for rule-based approach.

Suppose you have a dataset comprising of 100 images of

apples and 100 images of bananas. To train your machine,
you feed this data into the machine and label each image as
either apple or banana.

Now if you test the machine with the image of an apple, it

will compare the image with the trained data and according
to the labels of trained images, it will identify the test image
as an apple.
This is known as Rule-based approach. The rules given to
the machine in this example are the labels given to the
machine for each image in the training dataset. Observe
the following image:
 Learning Based

The machine is fed with data and the desired output

to which the machine designs its own algorithm (or
set of rules) to match the data to the desired output
fed into the machine to train.

In the learning-based approach, the relationship or

pattern in data is not defined by the developer.
This approach takes random data which is fed into the
machine and it is left to the machine to figure out the patterns
or required trends.

In general this approach is useful when the data is not

labelled and random for a human to use them.

Thus, the machine looks at the data, tries to extract similar

features out of it and clusters the same datasets together.

In the end as output, the machine tells us about the trends

which are observed in the training data.
For example, suppose you have a dataset of 1000 images of random
stray dogs of your area.

Now you do not have any clue as to what trend is being followed in this
dataset as you don’t know their breed, or colour or any other feature.

Thus, you would put this into a learning approach based AI machine
and the machine would come up with various patterns it has observed
in the features of these 1000 images.

It might cluster the data on the basis of colour, size, fur style, etc. It
might also come up with some very unusual clustering algorithm
which you might not have even thought of!
 AI PROJECT EVALUATION
Moving towards deploying the model in the real world,
we test it in as many ways as possible. The stage of testing
the models is known as EVALUATION.

Evaluation is a process of understanding the reliability of

any AI model, based on outputs by feeding the test
dataset into the model and comparing it with actual
answers.
Importance of Evaluation
Evaluation is a process that critically examines a
program. It involves collecting and analyzing
information about a program's activities,
characteristics, and outcomes. Its purpose is to
make judgments about a program, to improve its
effectiveness, and/or to inform programming
decisions.
• Evaluation is important to ensure that
the model is operating correctly and
optimally.
• Evaluation is an initiative to understand
how well it achieves its goals.
• Evaluations help to determine what
works well and what could be improved
in a program.
ONCE THE EVALUATION IS DONE IT MUST
BE DEPLOYED
AI PROJECT DEPLOYMENT
Deployment is the method by which you
integrate a machine learning model into an
existing production environment to make
practical business decisions based on data.
It is one of the last stages in the machine
learning life cycle and can be one of the
most cumbersome.
 JOB LOSS
With more machines being used for day-to-day work, the
fear of unemployment is increasing.
With the emergence of Al and automation, there will be
technology-driven societal changes.
The study reveals, depending various adoption scenarios,
automation will displace between 400 to 500 million jobs by
2030.
Al robots were replaced with Human employees in a
Japanese Henn-na Hotel. This hotel started with hospitality
robot staff in all the departments but with time the robots
could not provide efficient services to its customers and
were replaced with human employees after much struggle.
 PERSONAL PRIVACY
In today's scenarios we are surrounded by technology where an
individual's personal life can be tracked easily.
The gadgets and the apps used on a daily basis are Al-enabled. The
data gathering abilities of Al can access your data from the social
networking websites used by you.
The cameras installed use facial recognition to identify you in the
crowd. You are being followed and recorded everywhere without
information.
Apple’s Iphone X used advanced front-facing camera and machine
learning to create a 3-dimensional map of a face-for-face ID
recognition. The company claimed that it is programmed to work
without errors even for cosmetic changes.
 IF AI MAKES MISTAKES

Al can make mistakes. It's recorded that in 2016, when the

Uber company conducted a test on self-driving cars in San
Francisco, Uber's autonomous vehicle ran six red lights.
The situation got out of hand, but a licensed driver was
made to sit behind the wheel in case of emergency so he
took over the control of the situation immediately.
In another situation Microsoft's chatbot-Tay was released.
It learnt its language from the people on Twitter over time
that enabled it to have a meaningful conversation based on
a topic. But sadly it was taken offline within 16 hours of its
launch as it started tweeting randomly some abusive and
offensive content.
 AUTONOMOUS WEAPONS

Also known as "killer robots" can aim

independently by pre-programmed
instructions. Most of the technically advanced
countries are developing these autonomous
weapons to safeguard themselves. There are
many dangers in using these weapons.
 BLACK BOX PROBLEM

• Black box is any AI system whose inputs and operations aren't

visible to the user or another interested party.
• Black box AI models arrive at conclusions or decisions without
providing any explanations as to how they were reached.
• for example, an autonomous vehicle strikes a pedestrian when
we’d expect it to hit the brakes, the black box nature of the
system means we can’t trace the system’s thought process and
see why it made this decision.
AI BIAS
 DATA

AI systems are the result of the data that is

fed into them. The data used to train the AI
system is the first step to check for biasness.
The dataset for AI systems should be realistic
and need to be of a sufficient size.
 ALGORITHMS
The algorithms in itself do not add biasness to an AI
model, but it amplifies the biasness.
Example of an image classifier model. Its trained on
images in the public domain-pictures of people's
kitchens. It so happens that most of the images are
of women rather than men.
Al algorithms are designed to maximise the
accuracy. Therefore, an AI algorithm may decide
that the people in the kitchen are women despite
the fact some of the images are of men.
 PEOPLE

The last source of Al bias is people. Those

who design AI models focus on achieving the
desired goals On that path at times, the
biases of the developers are reflected in their
models. It's important to note here that the
ethics and Al bias are not the problems of the
machine but the humans behind the
machines.