Manav Rachna International School, sector -46,

gurugram

ARTIFICIAL
INTELIGENCE
PRACTICAL
RECORD

Name: Aditva Singh

Class: 11th pastuer
Session: 2024-25
Serial no. Topic name
01 Ai for all- certificate

02 AI for all- badge

AI for oceans Introductions

03
and certification

Supervised Learning
04 andReinforcement Learning
Activitythrough Artbot

Neural Network Concept

05
throughQuick Draw

06 AI Domain - Data

07 AI Domain - Computer Vision

08 CV - Shadow Art

09 AI Domain - NLP

Teachable Machine
Activity★ Image
10
Project★ Audio
Project★ Pose Project

11 Chatbots

12 IBM certification

AI ethics- survival of fittest

13
and moral machines

14 data visualization
1)AI for all certificates
1)AI for all Badges
 2)AI for Oceans
The "AI for Oceans" lesson from Code.org teaches students
the basics of machine learning, focusing on supervised
learning. In the activity, students help an AI bot clean the
ocean by classifying objects as "fish" or "not fish" and
labeling various sea creatures. This helps the AI learn to make
similar distinctions on its own using the provided training
data. The lesson also introduces the concept of AI bias,
highlighting the importance of diverse and representative
data to ensure fair and accurate AI predictions.

This activity employs Machine Learning especially Supervised

Learning where data is used to train a model to classify new
datasets

Machine Learning: is a subset of AI where a model is trained

with data which helps to work itself around data and develop
its own algorithms

Supervised Learning: where supervisions is required during

training of the AI model. The machine is trained with labeled
data(it is already tagged)
 AI biaseness:it is the issue where AI algorithm generate Biased
results which are due to incorrect assumptions made by developer
during training process.
Hour of code certificate
 4)Artbot
Art Bot - It is an interactive web-based game in which the quest is to find
and retrieve stolen art objects. We train the AI helper to recognise and
locate the objects hidden in a maze of dungeons. These employs
supervised learning and reinforcement learning.

Supervised Learning – It happens in the presence of a supervisor or a

teacher. The machine is trained with labelled data (which is already
tagged with correct answers). Then, the machine is prepared to predict
the outcomes of unforeseen or new data based on the learnt information
it gained during the training process.

Reinforcement Learning – It is an area of machine learning in which the

machine is not given examples of correct input-output
 5)Neural Network
Concept through Quick
Draw
Quick Draw is an engaging game that uses machine learning, specifically
neural networks, to recognize user-made doodles. As players contribute
their drawings, the system learns and improves its recognition capabilities.
The collected doodles form the world's largest dataset of its kind, which is
shared publicly to support advancements in machine learning research.
 6)AI Domain - Data
Data: A data domain encompasses a variety of values including raw facts,
numerical sets, characters, images, videos, and more. AI systems are
capable of handling vast amounts of data. Rather than merely processing
this data, they recognize patterns within it, enabling them to make
predictions based on similar data typessome online games like tic tac toe
and rock paper scissor are based on this domain.

Tic Tac Toe: It is a retro-version of the simple game of X’s and O’s
which we can play with the computer. It is based on the AI domain of Data.
It recognises the pattern in our moves and eventually starts defeating us
almost every time.
 Rock Paper Scissors: It is a classic rock-paper-scissors game against
AI. The computer is able to predict our moves based on our history because
we are very likely to repeat ourselves. It identifies patterns within the data.
The more we play, the more data is collected and the AI gets better at
predicting our moves.
 7)AI Domain - Computer
Vision
Computer Vision (CV): This is a branch of artificial
intelligence (AI) that enables machines to analyses visual
data, such as digital images and videos, extract meaningful
information and patterns, and ultimately make decisions,
take actions, or provide recommendations based on the
obtained information. Computer vision empowers machines
to perceive, observe, and comprehend the world around
them.Some online games based on Computer vision are
Teachable Snake and emoji scavenger hunt

Teachable Snake: Teachable Snake is an interactive web game where, instead of

using physical buttons, players control the snake with a black arrow drawn on
paper. By turning the paper in different directions in front of the webcam,
players can manoeuvre the snake through the game.
Emoji Scavenger Hunt: This online game challenges us to locate the
A neural network
will guess whatever comes in front of the camera.
 8)AI Domain- NLP
Natural Language Processing (NLP): This branch of AI
focuses on enabling computers to understand and interpret
spoken words or text as humans do. NLP combines
computational linguistics with statistical methods, machine
learning, and deep learning models. This allows computers
to process and comprehend human language in both text and
voice formats. Applications of NLP include Google
Assistant, language translators, sentiment analysis, and
more.Some games for example Luis and Semantris are built
using NLP.

Semantris: It is a word-association game that is powered by machine

learning. The AI is capable of doing so by being trained with billions of
online conversations. It predicts which words might come next in a
conversation correctly.
 Luis: It is a machine learning-based service to build natural language
into apps, bots, and IoT devices. It helps to create enterprise-ready, custom
models that constantly improve.
 9)Teachable Machines
Activity
Teachable Machine: Teachable Machine is a web-based tool that simplifies the
process of building machine learning models, making it fast and accessible to
everyone. This AI-based application allows users to train a computer to recognize
images, audio, and poses without writing any code. The process involves
gathering and grouping data, training the model, and then testing it to see if it
can accurately classify new examples. Once trained, the model can be exported
for use in various projects, such as websites and apps.

Image Project: A model can be taught to classify images using files

or a webcam with at least 60 images of the object.
Project link:- https://teachablemachine.withgoogle.com/models/lBA1DLT2-/

Audio Project: The model can be taught to classify audio by

recording short sound samples. At least 10 sound samples must be
recorded.
Project link:- https://teachablemachine.withgoogle.com/models/VbOgjGOe8/

Pose Project: The model can be taught to classify body positions

using files or striking poses in front of the webcam
project link:- https://teachablemachine.withgoogle.com/models/AI3-eKPPq/
 10)Chatbots
A chatbot (short for chatterbot) is a software program designed to simulate and
process human conversations through text or voice interactions, or both. It
enables humans to communicate with digital devices as if they were talking to a
real person. Chatbots can be integrated into and used through any major
messaging application.

Chat bots can be found almost anywhere some example are Chatgpt, Microsoft
copilot, Meta ai, Duolingo, Domino’s pizza bot etc.

Chatbots made
MedBot
https://poe.com/MedBot27

Finance_Bot

https://dashboard.sendbird.com/6398649D-0BEA-4B97-A96D-
023F19D3D216/ai-chatbots/preview/onboarding_bot
 PYTHON PRATICAL QUESTIONS
Q1. Write a Tipper program where the user inputs the total restaurant bill. The
program should then
display two amounts: 15 percent tip and 20 percent tip

Q2. Write a program that asks the user to enter their name and age. Print a
message addressed to the user that tells the user the year in which they will turn
100 years old.
Q3. Write the program to swap two numbers without using the third variables.

Q4. Write a program to accept the Basic salary from the user
and calculate the Net Salary. Net
Salary= Basic Salary + HRA + DA -PF , HRA=30% of Basic,
DA=20% of Basic, PF=12% of Basic.
Q5. Write a program to check whether the user is eligible for driving license
or not.

Q6. Your father always gives his car for service after 15000 km. Check whether
his car needs service or not. Read the kilometre reading from the user and give
the output.
 Q7.WAP to find the largest number among the three inputted numbers.

Q8. WAP to perform the following tasks according to the users’ choice using menu.
1)area of triangle 2) perimeter of square
3)perimeter of triangle 4) area of square
Q9. WAP to accept and display if the entered password full fill the following
criteria:
a) It must have at least 8 characters
b) Should have an uppercase
c) should have a digit
d) Should have a Special character

Q10. WAP that will arrange the three numbers entered in ascending or
descending order
Q11. Write a program to display the first ten even natural numbers

Q12.WAP to check if a string is a Palindrome or not.

Q13. WAP to print Fibonacci series’ first 200 elements

 Q14. Print the following pattern for n number of rows:

A)1 B)* C)1 D)4321

12 ** 13 432
123 *** 135 43
4

A)

B)

Your paragraph text

C)

D)
Q15. WAP to create an integer list and use the following function:
a) append () and extend ()
b) insert (), pop ()
c) count (), len ()
d) min (), max (), sum ()

A)

B)

C)

D)
IBM Skillsbuild certificate
 Aditva Singh

20 Jan 2025 (GMT)

Aditva Singh

19 January, 2025
 Aditva Singh

20 Jan 2025 (GMT)

 AI ethics
What are ethics in dimple terms?

Ethics : Ethics, in general, refers to the moral principles and values

that guide human behavior and decision-making. It involves
concepts such as distinguishing between right and wrong, promoting
fairness, upholding justice, and taking accountability for actions.

Ethics in AI : AI ethics focuses on the ethical considerations in

creating and using AI systems. It provides principles like fairness,
transparency, accountability, privacy, and minimizing harm to
ensure AI is designed and deployed responsibly, aligning with human
values for a positive societal impact.

Now About the five pillars of AI Ethics

Fairness Explanibility Robustness

Transparency Privacy
 Explanaibilty
Explainability in AI refers to the transparency and
interpretability of AI systems, allowing users to understand
how algorithms make decisions and predictions. It plays a
crucial role in ensuring that AI systems are transparent,
accountable, and aligned with ethical principles.
Essentially, explainability is about providing insights into
how AI models arrive at their decisions or predictions,
making their inner workings clear and understandable.

This clarity enables humans to follow the reasoning behind AI

outputs, verify their accuracy, and build trust in the system.
Explainability is fundamental for developing AI systems that are
not only high-performing but also ethical, reliable, and well-
suited for real-world applications where collaboration between
humans and AI is essential.
 Transparency
Transparency in AI involves making AI systems understandable
and accessible to users, stakeholders, and regulators. It includes
clear communication about how models work, the data they use,
and the processes behind their decision-making.

A key aspect of transparency is ensuring explanations

are accessible to non-technical users. This can involve
using visualizations, summaries, or intuitive language
to clarify complex AI processes.
 Fairness
Fairness in AI focuses on addressing and eliminating algorithmic
bias in machine learning models, particularly regarding sensitive
attributes like race, gender, age, sexual orientation, disability,
and socioeconomic status.

It ensures that AI systems are developed and used equitably,

transparently, and without discrimination. Fairness aims to
prevent harm by ensuring AI treats all groups fairly, avoiding
biases against protected characteristics.
 Robustness
Robustness in AI ensures that algorithms and systems function as
intended without unexpected errors or deviations. A robust AI
system is resilient, capable of handling errors, and effective in
real-world, unpredictable scenarios.

Techniques such as redundancy, error-checking, and noise-

handling help AI models process imperfect data while
maintaining reliable performance. Robustness is particularly
critical in high-stakes applications, where errors could lead to
financial losses, safety risks, or ethical issues.

The ultimate goal is to develop AI systems that are not only

effective but also resilient, reliable, and adaptable to diverse
real-world conditions.
 Privacy
Privacy refers to an individual's right to control their personal
information and be protected from unwanted intrusions. In the
context of AI, privacy ensures that sensitive data is safeguarded
throughout the development, deployment, and use of AI systems.

The misuse or mishandling of personal data can result in

privacy violations, identity theft, discrimination, and a loss of
trust in digital systems. Privacy in AI focuses on preventing
unauthorized access, protecting data from misuse, and
respecting users’ rights.
 BIAS, BIAS AWARENESS, SOURCES OF BIAS

Bias awareness involves recognizing that AI systems may exhibit

unfair preferences due to various factors such as the data they
are trained on, the algorithms they use, or the assumptions
underlying their design.

To address this critical issue, it is essential to remain vigilant

and proactively consider potential sources of bias throughout
the AI development process. This includes evaluating training
data for representativeness, ensuring fairness in algorithm
design, and continuously monitoring AI systems for unintended
bias in their outcomes.
 Types of biasness

Training Data Algorithimic

biasness biasness
AI systems make decisions by Algorithmic bias in AI results from
learning from training data, so it's both the algorithm's design and
crucial to evaluate datasets for any the programmer's decisions.
potential bias. One approach is to Algorithms may unintentionally
examine the data sampling for favor majority groups or reinforce
groups that may be inequalities, while programmers'
overrepresented or choices, like feature selection and
underrepresented in the training optimization, can reflect personal
data. biases.

Cognitive biasness
Cognitive bias refers to mental shortcuts that distort
judgment. Mitigation in AI involves identifying and
correcting biases in data, algorithms, or decisions
through diverse training data, regular audits, and
adjustments for fairness. Biases like confirmation and
availability bias help us process information quickly but
can skew our perception.
 Mitigating AI Biasness

Mitigation is the process of reducing the negative impact of

risks, problems, or biases. In AI, it involves identifying
biases or errors in data, algorithms, or decisions, and
applying strategies to address them, such as using diverse
training data, auditing models, adjusting algorithms for
fairness, and refining programmer practices to prevent
biased outcomes.

But why is it important?

Fairness and equality: Trust and Adoption :

in AI involve mitigating biases Reducing bias in AI builds
to ensure that systems treat trust among users and
all individuals and groups stakeholders. When people
fairly, preventing believe that AI systems are
discrimination and promoting fair and unbiased, they are
equal opportunities across more likely to adopt and rely
demographics.
on these technologies

Improved performance: Legal and ethical compliance:

Addressing biases can boost Mitigating biases helps
organizations adhere to legal
AI model performance.
regulations and ethical
Training on diverse and
standards. Addressing AI bias
representative data allows
can protect against legal
models to make more
liabilities and reputational
accurate predictions and
harm, as many regions have
decisions across a broader laws prohibiting
range of scenarios. discriminatory practices.
 Eradicating AI biasness

Using diverse data: Bias often arises from

unrepresentative data sets. Incorporating data that
reflects diversity in demographics, socioeconomic
backgrounds, and perspectives helps AI models
generalize better across populations and contexts.
Impact: A diverse data set prevents AI systems from
overfitting to specific groups and reduces the risk of
unintentional discrimination.

Detecting Bias Regularly auditing AI systems for bias is

crucial. Techniques such as fairness metrics,
comparing outcomes across demographic groups, and
testing against benchmarks for fairness can help
identify bias early in the model development process.
Impact : Bias detection allows teams to address
discrepancies before they negatively affect users.

Fair algorithms: Selecting or designing algorithms that

prioritize fairness helps reduce bias. Some algorithms
can adjust for imbalances or reweight data to produce
more balanced outputs.
Impact: Fair algorithms ensure that no group is unfairly
disadvantaged by the model.

Being transparent: Transparency in AI model design and

training fosters trust and accountability. Clear
explanations of data sources, model design, and
decision-making criteria help users understand and trust
AI systems.
Impact: Transparency makes it easier to identify and
address potential biases.
 Moral Machines
Moral Machines is an online game developed by the
Massachusetts Institute of Technology (MIT) to explore ethical
decision-making in AI systems. In the game, players are
presented with various moral dilemmas, such as deciding what
an autonomous car should do in an accident scenario. The game
collects data on players' choices to help researchers understand
how different cultures and individuals approach moral
decisions. The goal is to create a better understanding of ethical
preferences for use in designing AI systems that make morally
informed decisions.
Survival of the best fit
Survival of the Best Fit is an educational game focused on AI
hiring bias. It explores how AI systems can inherit human biases
and perpetuate inequality, emphasizing the importance of
understanding and addressing algorithmic bias. The game aims
to empower players by highlighting the need for accountability
in AI development and the risks of misusing technology. It is
designed to raise awareness about AI's impact on society and
the potential for improving its fairness.
 Data Visualization

Data visualization is crucial for exploring datasets before

applying AI models. By using tools like scatter plots,
histograms, or line graphs, you can identify trends,
patterns, and outliers that might not be evident in raw
data. Visualizing data helps to inform decisions and
improve the understanding of underlying structures, which
is essential for accurate AI model building.

Code 1 : Plot a line chart representing the weekly number of customer inquiries
received by a customer service center

Code output
 Code 2 : Visualize the distribution of different types of transportation used by
commuters in a city using a pie chart

Code output

Code 3:Plot a scatter plot to visualize the relationship between the height and weight of
grade 11 students.

Code output
 Code 4: A teacher wants to visualize the average marks obtained by Grade 11
students in various subjects.

Code output

Code 5:A teacher has collected the following data on the heights (in centimeters) of 30
students in Class 11.
Create a histogram to visualize the distribution of heights among the students. Use an
appropriate number of 6 bins to represent the data effectively.

Code output