Course Name - Artificial Intelligence

Course Code - SHIT601

Semester – IV

Dr. Rajani, IBS Bangalore

1
Artificial Intelligence – SHIT601 - Syllabus
• Introduction to the course, What is AI, Fundamentals of AI, usage of AI in a Business Perspective
• Basic Programming and other Data structural (DS) concepts for AI
• Agents and Environments, Good Behaviour: The Concept of Rationality, the Nature of Environments, Structure of Agents.
• Constraint Satisfaction Problems, Backtracking Search for CSPs. Local Search for Constraint Satisfaction Problems, The
Structure of Problems.
• To Study Informed Search and Exploration: Heuristic Function, Hill Climbing Function, 8 queen problem
• To Study First order logic: Representation Revisited Syntax and Semantic of First-order Logic, Models for first order logic,
Symbols and interpretations Atomic Sentences Complex Sentences Quantifiers
• Inference in First-order Logic Forward and Backward Chaining
• Knowledge Representation
• Uncertain Knowledge and reasoning
• Probability Reasoning
• Deep Learning, Q Learning and applications, Neural Networks, Optimizing Processes

2
Resources for the course- Books

Recommended Text Book:

• Artificial Intelligence: Making a System Intelligent, Dr. Nilakshi Jain, Wiley

Suggested Reference Books:

• Artificial Intelligence A Modern Approach – Second Edition, Stuart Russell, Peter
Norvig, Prentice Hall
• Artificial Intelligence - Second Edition – Elaine Rich, Kevin Knight, TATA McGraw-Hill.
• Artificial Intelligence - SK Bansal, 2012

3
Expected Learning Outcomes:

At the end of the course, the student is expected to

•Understand the significance of Artificial Intelligence.

•Understand some of the techniques used in AI to derive solutions to business and/or
real time problems
•Realize the use of AI in optimizing business processes.

4
Evaluation Timelines and Weightage

Component Component Number Expected slot / due date Weightage

Class Participation 1 Session 33 10
Test 1 Session 12 20
Test 2 Session 22 20
Test /Seminar 3 Session 32 10
End Term 4 After session 33 40

5
Introduction to AI

Artificial Intelligence is composed of two words Artificial and Intelligence, where

Artificial defines "man-made," and intelligence defines "thinking power", hence AI
means "a man-made thinking power.“

Definition
"It is a branch of computer science by which we can create intelligent machines which
can behave like a human, think like humans, and able to make decisions."

6
Deep Learning

Deep learning is a subset of machine learning, which is essentially a neural network with multiple
layers. These neural networks attempt to simulate the behaviour of the human brain.

Deep learning is a branch of machine learning which is based on artificial neural networks. It
is capable of learning complex patterns and relationships within data. It is designed to learn
from large amounts of data.

ML is best for well-defined tasks with structured and labelled data.

Deep learning is best for complex tasks that require machines to make sense of unstructured data.

7
Basic Concept of AI, ML and DL

8
Machine Learning

• Supervised Learning: Start with the basics of supervised learning, where the
algorithm is trained on labeled datasets to make predictions.

• Unsupervised Learning: Explore unsupervised learning, where the algorithm

identifies patterns and structures in data without labeled examples.

9
Deep Learning architectures include

Convolutional Neural Networks (CNNs) - are specifically for image and video recognition
tasks. CNNs can automatically learn features from the images, which makes them well-suited
for tasks such as image classification, object detection, and image segmentation.

Recurrent Neural Networks (RNNs) - are a type of neural network that can process sequential
data, such as time series and natural language.

10
Relationship - AI, ML, NN and DS

11
Applications of Deep Learning :

Computer Vision - Deep learning models can enable machines to identify and understand
visual data. Some of the main applications of deep learning in computer vision include:

•Object detection and recognition: A deep learning model can be used to identify and
locate objects within images and videos, making it possible for machines to perform
tasks such as self-driving cars, surveillance, and robotics.

•Image classification: Deep learning models can be used to classify images such as
animals, plants, and buildings. This is used in applications such as medical imaging,
quality control, and image retrieval.

12
Natural language processing (NLP):

The Deep learning model can enable machines to understand and generate human language. Some of the
main applications of deep learning in NLP include:

•Automatic Text Generation – A deep learning model can learn the corpus of text and new text like
summaries, and essays can be automatically generated using these trained models.

•Language translation: Deep learning models can translate text from one language to another, making it
possible to communicate with people from different linguistic backgrounds.

•Sentiment analysis: Deep learning models can analyse the sentiment of a piece of text, making it possible
to determine whether the text is positive, negative, or neutral. This is used in applications such as customer
service, social media monitoring, and political analysis.

•Speech recognition: Deep learning models can recognize and transcribe spoken words, making it possible
to perform tasks such as speech-to-text conversion, voice search, and voice-controlled devices.

13
ANN:-

An Artificial neural network is usually a computational network based on biological

neural networks that construct the structure of the human brain.

Similar to a human brain has neurons interconnected to each other, artificial neural
networks also have neurons that are linked to each other in various layers of the
networks.

Neurons are called as nodes.

14
Artificial Neural Network looks like:-
The artificial neural network takes input and computes the
weighted sum of the inputs and includes a bias. This
computation is represented in the form of a transfer function.

It determines weighted total is passed as an input to an

activation function to produce the output. Activation
functions choose whether a node should fire or not. Only
those who are fired make it to the output layer.
Weight and Bias in Neural Network
Neurons are the basic units of a neural network. In an ANN, each neuron in a layer is connected to
some or all of the neurons in the next layer. When the inputs are transmitted between neurons, the
weights are applied to the inputs along with the bias.

Weights control the signal between two neurons. In other words, a weight decides how much
influence the input will have on the output.

Biases, which are constant, are an additional input into the next layer that will always have the value of
1. Bias units are not influenced by the previous layer but they do have outgoing connections with their
own weights. The bias unit guarantees that even when all the inputs are zeros there will still be an
activation in the neuron.

Activation Function is a mathematical formula which helps the neuron to switch ON/OFF.
Each artificial neuron has the following main functions:

•Takes inputs from the input layer

•Weighs them separately and sums them up
•Pass this sum through a nonlinear function to produce output.
Perceptron Function

Perceptron is a function that maps its input “x,” which is multiplied with the learned weight coefficient;
an output value ”f(x)”is generated.

“w” = vector of real-valued weights

“b” = bias (an element that adjusts the boundary away from origin without any dependence on the
input value)
“x” = vector of input x values

The output can be represented as “1” or “0.” It can also be represented as “1” or “-1” depending on
which activation function is used.
Consider a neuron with two inputs (x1,x2) as
The values of the two inputs(x1,x2) are 0.8 and 1.2
We have a set of weights (1.0,0.75) corresponding to the two inputs
Then we have a bias with value 0.5 which needs to be added to the sum
 The input to activation function is then calculated using the
formula:-

Now the combination(C) can be fed to the activation function.

In our case, the combination value we got was 2.2 which is greater than 0 so the
output value of our activation function will be 2.2.
This will be the final output value of our single-layer neuron.
 Architecture of
Artificial Neural Networks
 Architecture of
Artificial Neural Networks

This neural network is formed in three layers, called the input layer, hidden layer, and output
layer.

Each layer consists of one or more nodes, represented in this diagram by the small circles.

The lines between the nodes indicate the flow of information from one node to the next.

The nodes of the input layer are passive, means they do not modify the data. They receive a single
value on their input, and duplicate the value to their multiple outputs. In comparison, the nodes of
the hidden and output layer are active.
Connections

Nodes
Each value from the input layer is duplicated and sent to all of
the hidden nodes. This is called a fully
interconnected structure.

The values entering a hidden node are multiplied by weights.

The weighted inputs are then added to produce a single

number using ∑.

This number is passed through a nonlinear mathematical

function called a sigmoid.
Artificial Neural Network consists of three layers:

Input Layer:
It accepts inputs in several different formats provided by the programmer.

Hidden Layer:
The hidden layer presents in-between input and output layers. It performs all the
calculations to find hidden features and patterns.

Output Layer:
The input goes through a series of transformations using the hidden layer, which
finally results in output that is conveyed using this layer.
In this particular type of neural network, the information flows only from the input to the
output (from left-to-right).

Other types of neural networks have more intricate connections, such as feedback paths.
 Models of neural networks:

Feed-forward neural network

Feedback neural network (recurrent or interactive)

 Feed forward neural network

Feed-forward ANNs allow signals to travel one way only: from input to output.
There is no feedback or loops. The output of any layer does not affect that same layer in such
networks.
Feed-forward ANNs tend to be straightforward networks that associate inputs with outputs.
This model is extensively used in pattern recognition. Also referred to as bottom-up or top-down.
They have fixed inputs and outputs. They are mostly used in pattern generation, pattern
recognition and classification.
A threshold transfer function is sometimes used to quantify the output of a neuron in the output
layer.
Feed-forward networks include Perceptron (linear and non-linear).
Feed-forward networks are often used in data mining.
Feedforward neural networks are ideally suitable for modelling relationships between a set of
predictor or input variables and one or more response or output variables.

In other words, they are appropriate for any functional mapping problem where we want to
know how a number of input variables affect the output variable.

The multilayer feed-forward neural networks, also called multi-layer perceptron (MLP), are the
most widely studied and used neural network model in practice.
Feed forward neural network
 Feedback neural network

Signals can travel in both the directions in Feedback neural networks. Feedback neural
networks are very powerful and can get very complicated.

Feedback neural networks are dynamic. The ‘state’ in such network keep changing until
they reach an balance point.
Feedback neural network
 Perceptron (Artificial Neuron)
Based on linear threshold unit (LTU)

Perceptron is the basic unit of the neural network. In simple terms, it is a mathematical function based on
a model of biological neurons. It can also be seen as a simple logic gate with binary outputs.

A Perceptron is an algorithm for supervised learning of binary classifiers.

The Perceptron receives multiple input signals, and if the sum of the input signals exceeds a certain
threshold, it either outputs a signal or does not return an output. In the context of supervised learning
and classification, this can then be used to predict the class of a sample.

The threshold is one of the key components of the perceptron. It determines, based on the inputs,
whether the perceptron fires or not.
 Types of Perceptron

ØSingle layer Perceptron

ØMultiple layer Perceptron

 Single layer Perceptron
Single-layer perceptron can learn only linearly separable patterns.

A single-layer perceptron (SLP) is a feed-forward network based on a threshold transfer function.

SLP is the simplest type of artificial neural network and can only classify linearly separable
cases with a binary target (1 , 0).

The single-layer perceptron does not have a priori knowledge, so the initial weights are assigned
randomly.

SLP sums all the weighted inputs and if the sum is above the threshold (some predetermined
value), SLP is said to be activated (output=1).
The input values are presented to the perceptron, and if the predicted
output is the same as the desired output, then the performance is
considered satisfactory and no changes to the weights are made.
However, if the output does not match the desired output, then the
weights need to be changed to reduce the error.
 Multiple layer Perceptron - Backpropagation algorithm
Multilayer Perceptron or feed-forward neural networks with two or more layers have the greater
processing power. A multilayer perceptron is a type of feed-forward artificial neural network that
generates a set of outputs from a set of inputs. An MLP is a neural network connecting multiple
layers in a directed graph, which means that the signal path through the nodes only goes one
way.
Multi-layer Perceptron – Back propagation algorithm

A multi-layer perceptron (MLP) has the same structure as a single-layer perceptron with one or
more hidden layers.

The backpropagation algorithm consists of two phases:

•the forward phase where the activations are propagated from the input to the output layer,

•and the backward phase, where the error between the observed actual and the requested
nominal value in the output layer is propagated backwards in order to modify the weights and
bias values.
 Data at input layer is multiplied with weights to form hidden layer
h1 = (x1 * w1) + (x2 * w1)
h2 = (x1 * w2) + (x2 * w2)
h3 = (x1 * w3) + (x2 * w3)

Output of hidden layer is passed through a non-linear function also known as

activation function to form guessed output

y = fn( h1 , h2, h3 )

MLP networks are used for supervised learning format. A typical learning
algorithm for MLP networks is also called back propagation's algorithm.
 Need of Back propagation ?

While designing a Neural Network, in the beginning, we initialize weights with some
random values or any variable for that fact.

It’s not necessary that whatever weight values we have selected will be correct, or it fits
our model the best.

how will you reduce the error?

 Back propagation:
One way to train our model is called as Back propagation.

Back propagation is a supervised learning algorithm, for training Multi-layer Perceptron

(Artificial Neural Networks).

The Back propagation algorithm looks for the minimum value of the error function in weight
space using any learning rule.

The weights that minimize the error function are then considered to be a solution to the
learning problem.
 Backpropagation algorithm

44

1.Initialization:
1. Initialize the weights and biases of the neural network randomly or using a
specific initialization strategy.

2.Forward Pass:
1. Input data is fed forward through the network to generate a predicted output.
2. The forward pass involves passing the input through each layer of the network,
applying weights, biases, activation functions, and producing the final output.
 45
3. Compute Loss:
1. Calculate the difference (error or loss) between the predicted output and the actual target
values using a loss function.
2. Common loss functions include Mean Squared Error (MSE) for regression tasks or Cross-
Entropy Loss for classification tasks.
4. Backward Pass (Backpropagation proper):
1. Compute the gradients of the loss with respect to the weights and biases using the chain
rule of calculus.
2. Starting from the output layer and moving backward through the network, calculate the
partial derivatives of the loss with respect to each parameter.
 46

5. Update Weights and Biases:

1. Use the computed gradients to update the weights and biases of the network in the
opposite direction of the gradient.
2. The update rule typically involves multiplying the gradient by a learning rate and
subtracting it from the current weights and biases.
 Q-learning- Reinforcement learning
47
Q-learning is a machine learning approach that enables a model to iteratively learn and
improve over time by taking the correct action.

Q-learning is a reinforcement learning technique used for making decisions in an

environment with an agent that takes actions to achieve goals. It is a model-free
algorithm, meaning it does not require knowledge about the environment's dynamics
in advance.
 Q-learning- Reinforcement learning
48

A common drawback of Supervised learning is the vast amount of data that models need
to train.

Reinforcement Learning overcomes the problem of data acquisition by almost

completely removing the need for data.

Reinforcement learning is a branch of Machine Learning that trains a model to come to

an optimum solution for a problem by taking decisions by itself.
 Steps on how Q-learning is introduced:

49

1. Understanding Reinforcement Learning

2. Markov Decision Processes (MDPs)
3. Reward Systems
4. Q-values and Q-learning Basics
5. Exploration and Exploitation
6. Q-learning Algorithm
7. Convergence Conditions
 Q-learning- Reinforcement learning
50
It consists of:

•An Environment, which an agent will interact with, to learn to reach a goal or perform an
action.

•A Reward if the action performed by the model is bringing us closer to the goal/is leading to
the goal. This is done to train the model in the right direction.

•A negative reward if it performs an action that will not lead to the goal to prevent it from
learning in the wrong direction.
 51
Q-Learning is a Reinforcement learning policy that will find the next best action, given a current state.
It chooses this action at random and aims to maximize the reward.

Important Terms in Q-Learning

1.States: The State, S, represents the current position of an agent in an environment.

2.Action: The Action, A, is the step taken by the agent when it is in a particular state.
3.Rewards: For every action, the agent will get a positive or negative reward.
4.Episodes: When an agent ends up in a terminating state and can’t take a new action.
5.Q-Values: Used to determine how good an Action, A, taken at a particular state, S, is. Q (A, S).
6.Temporal Difference: A formula used to find the Q-value by using the value of current state and action and previous
state and action.
 Markov Decision Process

52
Markov Decision Process (MDP), is used to formalize the
reinforcement learning problems.

In MDP, the agent constantly interacts with the environment and

performs actions; at each action, the environment responds and
generates a new state.

A Markov model is a stochastic method for randomly changing systems

that possess the Markov property. This means that, at any given time,
the next state is only dependent on the current state and is
independent of anything in the past.
 Markov Decision Process
53

MDP is used for decision-making problems where the outcomes are partly random and
controllable.

MDP contains a tuple of four elements (S, A, Pa, Ra):

•A set of finite States S

•A set of finite Actions A
•Rewards received after transitioning from state S to state S', due to action a.
•Probability Pa.
 The architecture of an MDP can be summarized as follows:

54

•Agent: The decision-maker that interacts with the environment.

•Environment: The external system that the agent interacts with, and where state transitions
occur based on agent actions.
•State Space: The set of all possible states the system can be in.
•Action Space: The set of all possible actions the agent can take.
•Transition Model: Describes how the system transitions from one state to another based on the
agent's actions.
•Reward Model: Specifies the immediate rewards associated with state-action pairs.
•Policy: The strategy or mapping that guides the agent's decision-making.
Example – Action of robot
55
 Reward Systems
56

A reward function is a crucial component of reinforcement learning that enables agents to

learn from their own actions and feedback.

In reinforcement learning, rewards are associated with actions to provide feedback to the
learning agent about the desirability of its actions.

By associating rewards with actions, the agent can learn which actions lead to favourable
outcomes and which do not.
 Discount Factor - balances immediate rewards against future rewards.

57

It is used to balance the importance of immediate rewards against future rewards. It

influences the agent's decision-making process and affects how much weight is given
to future outcomes.

The discount factor is typically a value between 0 and 1, with 0 indicating that the
agent only considers immediate rewards and 1 indicating that the agent values future
rewards equally to immediate rewards.
 Q-values and Q-learning
58

Q-values is the metric used to measure an action at a particular state.

Methods to determine the Q-value:

Temporal difference. The temporal difference formula calculates the Q-value by incorporating
the value of the current state and action by comparing the differences with the previous state and
action.

Q-learning is a popular algorithm that utilizes Q-values to learn an optimal policy in a model-
free manner, updating its estimates based on observed rewards and future Q-values.
 Exploration and Exploitation
59
Exploration is any action that lets the agent discover new features about the environment, while
exploitation is capitalizing on knowledge already gained.

Exploitation is defined as a greedy approach in which agents try to get more rewards by using estimated
value but not the actual value. So, in this technique, agents make the best decision based on current
information.

In exploration, agents primarily focus on improving their knowledge about each action instead of getting
more rewards so that they can get long-term benefits. So, in this technique, agents work on gathering
more information to make the best overall decision.
 Q-learning algorithm process includes the following
60
•Q-table initialization. The first step is to create the Q-table as a place to track each action in
each state and the associated progress.
•Observation. The agent needs to observe the current state of the environment.
•Action. The agent chooses to act in the environment. Upon completion of the action, the model
observes if the action is beneficial in the environment.
•Update. After the action has been taken, it's time to update the Q-table with the results.
•Repeat. Repeat steps 2-4 until the model reaches a termination state for a desired objective.
 Convergence Conditions
61

Convergence refers to the point at which the training process reaches a stable state and
the parameters of the network (i.e., the weights and biases) have settled on values that
produce accurate predictions for the training data.

A neural network can be considered to have converged when the training error (or loss)
stops decreasing or has reached a minimum level of acceptable error.
 Optimizing Process through Q learning –
find optimal strategies for decision-making in dynamic environments

62

Process optimization is the systematic approach of analysing and improving business

processes to achieve maximum efficiency, effectiveness, and quality. The goal of process
optimization is to continuously reduce waste, increase productivity, and ultimately improve
an organization's bottom line.

Q-learning can be applied to optimize processes in various domains by enabling an agent to

learn a policy that maximizes cumulative rewards. The algorithm is particularly well-suited
for problems where an agent interacts with an environment, takes actions, and receives
feedback in the form of rewards
 Example
63
Resource Allocation and Management:

•Inventory Management: Q-learning can help optimize inventory management by

learning policies that minimize holding costs, stockouts, and other relevant factors.

•Energy Management: In smart grids or energy systems, Q-learning can optimize the
control of devices to manage energy consumption efficiently.
 Q-learning can be applied to optimize processes in AI - Steps
64

1. Define the Problem as a Markov Decision Process (MDP):

• States, Actions, and Rewards: Clearly define the states of the system, the
possible actions an agent can take in each state, and the immediate rewards
associated with each action-state pair.
 Q-learning can be applied to optimize processes in AI - Steps
65

2. Formulate the Q-learning Objective:

• Objective Function: Express the goal of Q-learning as learning the optimal Q-

values, representing the expected cumulative rewards for each state-action pair.
 Q-learning can be applied to optimize processes in AI - Steps
66

3. Design the Q-learning Algorithm:

• Initialization: Initialize Q-values for all state-action pairs.

• Exploration-Exploitation: Implement a strategy for balancing exploration (trying new actions) and
exploitation (choosing actions with known high rewards).
• Q-value Update Rule: Use the Q-learning update rule to iteratively update Q-values based on
observed rewards and estimated future rewards.
• Learning Rate: Tune the learning rate to control the impact of new information on the Q-values.
• Discount Factor: Adjust the discount factor to balance immediate rewards against future rewards.
 Q-learning can be applied to optimize processes in AI - Steps
67

4. Implementation of Q-learning:

• Coding: Implement the Q-learning algorithm in a programming language, making

use of data structures to represent states, actions, and Q-values.
• Environment Interaction: Set up the interaction between the agent and the
environment, allowing the agent to take actions, receive rewards, and update Q-
values.
 Q-learning can be applied to optimize processes in AI - Steps
68

5. Experimentation and Fine-Tuning:

• Hyperparameter Tuning: Experiment with different values for hyperparameters,

including the learning rate and discount factor, to find the optimal configuration.
• Monitoring Convergence: Monitor the convergence of Q-values over time to ensure
that the algorithm is learning effectively.
 Q-learning can be applied to optimize processes in AI - Steps
69

6. Validation and Testing:

• Validation Set: Use a validation set to evaluate the performance of the learned
policy during the training process.
• Test Set: Assess the final Q-values on a separate test set to ensure the
generalization of the learned policy to new situations.
 Q-learning can be applied to optimize processes in AI - Steps
70
7. Scale Up to Real-World Processes:

• Integration with Real Systems: Adapt the Q-learning approach to real-world

processes by integrating it with the actual systems and processes.

• Sensitivity Analysis: Perform sensitivity analysis to understand how changes in

the environment or system dynamics impact the learned policy.
 Q-learning can be applied to optimize processes in AI - Steps
71

8. Continuous Improvement:

• Reinforcement Learning Iterations: Apply the lessons learned from the initial
Q-learning implementation to iterate and improve the model continuously.
• Dynamic Environments: Adjust the Q-learning approach to handle dynamic
environments where the underlying system may change over time.
 Q-learning can be applied to optimize processes in AI - Steps
72

9. Monitoring and Maintenance:

• Monitoring Performance: Continuously monitor the performance of the Q-

learning model in the deployed system.
• Adaptation: Implement mechanisms for the model to adapt to changes in the
environment or system requirements.
 Example - Elevator control system
Designing an elevator control system using Q-learning can lead to more efficient and
optimised elevator operations, especially in multi-floor buildings.

73
1. State Representation:

Define the state of the elevator system by considering factors such as the current floor of
the elevator, the direction it is moving, the current load (number of passengers), and the
historical patterns of elevator usage.

2. Action Representation:
•Actions can be represented by the different commands the elevator can execute,
including moving up, moving down, stopping, and opening/closing doors. The control
system needs to decide which action to take based on the current state.
 74

3. Q-Table Initialization:
•Initialize a Q-table with entries for each state-action pair. The Q-values represent the
expected cumulative reward for taking a particular action in a given state. Initially, these
values can be set to zero.
4. Reward Structure:
•Design a reward structure to encourage behaviours that minimize waiting times and energy
consumption. Positive rewards can be given for prompt arrivals at requested floors and
efficient use of energy, while negative rewards can be assigned for delays and excessive
energy usage.
 75

5. Learning Algorithm:
•Utilize the Q-learning update rule to iteratively update the Q-values based on observed
rewards and future expected rewards. This involves adjusting the Q-values to learn optimal
strategies over time.
6. Exploration-Exploitation Strategy:
•Implement an exploration-exploitation strategy, such as epsilon-greedy, to balance between
exploring new actions and exploiting learned optimal actions. This ensures the system adapts
to changing passenger demands and traffic patterns.
 76
7. Energy-Efficient Movement:
•Integrate algorithms for energy-efficient movement, such as regenerative braking and
optimizing the use of the elevator's motor. Q-learning can guide the elevator to make
decisions that minimize unnecessary energy consumption.

8. Dynamic Load Balancing:

•Consider dynamic load balancing to distribute passengers among available elevators
efficiently. Q-learning can help in making decisions that optimize the distribution of
passengers to minimize waiting times.
 77
9. Real-Time Adaptation:
•Allow the system to adapt in real-time to changing passenger demands, building traffic,
and any mechanical issues. Q-learning enables continuous learning and adjustment to
optimize performance.

10. Testing and Validation:

•Implement simulation environments to thoroughly test and validate the Q-learning-based
elevator control system. Evaluate its performance under various scenarios, including peak
hours, emergency situations, and changes in building occupancy.
 Assignment
Regular Assessment – 4 Marks
78
Games
Rock Paper Scissor
Snake game
Tic Tac Toe game
Fruit ninja game, also known as a fruit-slicing game is easy to play.

To-do list
The chat bot on a specific topic
Calculator
Other similar task
Thank You

79