You are on page 1of 16

A Synopsis Report

On
Personalized secured framework for exam preparation using AI
Submitted to
DELHI TECHNICAL CAMPUS
(Affiliated Guru Gobind Singh Indraprastha University, New Delhi)

Greater Noida

in partial fulfillment of the requirements for the award of the degree of


Bachelor of Technology
by

Anuj Chhabra (02318002720)


Abhinav Kumar (00518002720)
Deepanshu Sardana (04018002720)
Akshit Kumar Dhaka (01118002720)

under the guidance of

Ms. Upasna Joshi

COMPUTER SCIENCE AND ENGINEERING


DELHI TECHNICAL CAMPUS
GREATER NOIDA (U.P.)

October 2023
CONTENTS

1. Problem Statements
1
2. Objectives
2
3. Introduction
3 -6
3.1 Feasibility Study
4
3.2 Need and Significance
5
3.3 Intended user
6
4. Literature Review
7
5. Proposed Methodology in brief
8-12
5.1 Functional Requirements
8
5.2 Non-Functional Requirements
9
5.2.1 Usability
9
5.2.2 Availability
9
5.2.3 Efficiency
10
5.2.4 Accuracy
10
5.2.5 Performance
10
5.2.6 Reliability
10
5.2.7 Maintainability
10
5.2.8 Security
10
5.3 Hardware Requirements
10
5.4 Software Requirements
11
5.4.1 Front End
11
5.4.2 Back End
11
6. Diagrams
12
6.1 Use Case
12
7. References
13
1. Problem Statement

Traditional exam-taking methods are often inefficient, time-consuming, and costly.


Additionally, they can be difficult to scale and secure. As a result, there is a need for a more
modern and effective way to take exams.

A digital exam-taking website with AI-generated questions could address many of the
challenges of traditional exam-taking methods. Such a system could offer the following benefits:

Lack of personalization: Traditional exam preparation methods often fail to personalize the
learning experience for each student. As a result, students may focus on areas where they
are already strong and neglect areas where they need the most practice.
Lack of variety: Traditional exam preparation methods often provide students with a
limited variety of practice questions. This can make it difficult for students to prepare for
different types of questions that they may encounter on the exam.
Lack of feedback: Traditional exam preparation methods often fail to provide students with
immediate feedback on their answers. This can make it difficult for students to identify their
strengths and weaknesses and to track their progress over time.
Lack of convenience: Traditional exam preparation methods often require students to study
from textbooks and take practice tests in a physical location. This can be inconvenient for
students who have busy schedules or who live in remote areas.

However, there are also some challenges that need to be addressed in order to develop and
implement a successful digital exam-taking website with AI-generated questions. These
challenges include:

Bias: AI-generated questions need to be carefully designed to avoid bias against certain
groups of test-takers.
Accuracy: AI-generated questions need to be accurate and up-to-date.
Fairness: AI-generated questions need to be fair and equitable for all test-takers.
Security: AI-generated questions need to be secure to prevent cheating and plagiarism.

Despite these challenges, a digital exam-taking website with AI-generated questions has the
potential to revolutionize the way that exams are taken. By addressing the challenges of
traditional exam-taking methods, such as efficiency, scalability, security, and personalization,
AI-generated exams can help to improve the exam-taking experience for both test-takers and
test-makers.

1
2. Objectives

The objectives for a digital exam-taking website with AI-generated questions could include:

Improve the efficiency and scalability of exam-taking: AI-generated questions can be created
and delivered much more quickly and efficiently than traditional paper-and-pencil exams.
This can save time and money for both test-takers and test-makers. Additionally, a digital
exam-taking website can be easily scaled to accommodate large numbers of test-takers.
Enhance the security of exam-taking: A digital exam-taking website can be made more
secure than traditional paper-and-pencil exams using a variety of security measures, such as
encryption and user authentication. This can help to prevent cheating and plagiarism.
Personalize the exam-taking experience: AI-generated questions can be personalized to the
individual test-taker's skill level and learning needs. This can help to ensure that all test-
takers have a fair and equitable opportunity to succeed.
Provide a more engaging and interactive exam-taking experience: A digital exam-taking
website can incorporate multimedia elements, such as images, videos, and simulations, to
make the exam-taking experience more engaging and interactive for test-takers.
Improve the quality of feedback provided to test-takers: A digital exam-taking website can
provide test-takers with immediate feedback on their answers. This feedback can help test-
takers to identify their strengths and weaknesses and to improve their learning.

In addition to these general objectives, a digital exam-taking website with AI-generated


questions could also be designed to meet specific needs, such as:

Providing a secure and reliable platform for high-stakes exams: A digital exam-taking website
could be used to administer high-stakes exams, such as college entrance exams and
professional certification exams.
Supporting formative assessment: A digital exam-taking website could be used to provide
formative assessment to students throughout the learning process. This feedback can help
students to identify and address their learning gaps early on.
Accommodating students with disabilities: A digital exam-taking website could be designed
to accommodate students with disabilities by providing them with additional time, features,
or support.

Overall, the objectives of a digital exam-taking website with AI-generated questions should be
to improve the efficiency, security, personalization, engagement, and feedback provided to test-
takers.

2
3. Introduction

A digital exam preparation website with AI-generated questions is a website that helps students
prepare for exams by providing them with personalized practice questions. The website uses
artificial intelligence (AI) to generate questions that are tailored to the individual student's skill
level and learning needs.

This type of website offers a number of advantages over traditional exam preparation methods,
including:
Personalization: AI-generated questions can be personalized to the individual student's skill
level and learning needs. This helps students to focus on the areas where they need the most
practice.
Variety: AI-generated questions can be generated in a variety of formats, including multiple
choice, fill-in-the-blank, and essay questions. This helps students to practice answering
different types of questions that they may encounter on the exam.
Feedback: AI-generated questions can provide students with immediate feedback on their
answers. This helps students to identify their strengths and weaknesses and to track their
progress over time.
Convenience: Digital exam preparation websites can be accessed from anywhere with an
internet connection. This makes it easy for students to study at their own pace and on their
own schedule.

Overall, a digital exam preparation website with AI-generated questions is a valuable tool for
students who are preparing for exams. By providing personalized practice questions, immediate
feedback, and convenience, this type of website can help students to improve their scores and to
achieve their academic goals.

Here is an example of how a digital exam preparation website with AI-generated questions could
be used:
A student is preparing for a history exam. They visit a digital exam preparation website and create
an account. The website then asks the student a few questions about their skill level and learning
needs. Based on this information, the website generates a personalized set of practice questions for
the student. The student can then start practicing the questions and receive immediate feedback on
their answers. The student can continue to practice until they feel confident about their knowledge
of the material.

Digital exam preparation websites with AI-generated questions are a valuable resource for
students who are preparing for exams. By providing personalized practice questions, immediate
feedback, and convenience, this type of website can help students to improve their scores and to
achieve their academic goals.

3
, Feasibility Study
3.1

Technical Feasibility:-
The technology required to develop and implement a digital exam preparation website with AI-
generated questions is already available. There are a number of cloud-based platforms that can
be used to develop and deploy such a website. Additionally, there are a number of AI-powered
question generation tools that can be used to generate personalized practice questions for
students.

Some of the technical challenges that need to be addressed include:


Accuracy: AI-generated questions need to be accurate and up-to-date. This can be achieved
by using a variety of techniques, such as human review and natural language processing.
Fairness: AI-generated questions need to be fair and equitable for all students. This can be
achieved by using a variety of techniques, such as bias mitigation and diversity testing.
Security: AI-generated questions need to be secure to prevent cheating and plagiarism. This
can be achieved by using a variety of techniques, such as encryption and user
authentication.

Financial Feasibility:-
The financial feasibility of a digital exam preparation website with AI-generated questions
depends on a number of factors, including the development costs, marketing costs, and revenue
model.

The development costs for such a website would vary depending on the complexity of the
website and the features that are included. However, it is estimated that the development costs
for a basic website would be in the range of $50,000 to $100,000.

The marketing costs for such a website would also vary depending on the marketing channels
that are used. However, it is estimated that the marketing costs would be in the range of $10,000
to $20,000 per month.

The revenue model for such a website would typically be based on a subscription model.
Students would pay a monthly or annual subscription fee to access the website and its features.
The subscription fee would need to be set high enough to cover the development costs,
marketing costs, and operating costs of the website.

Security:-
Security is a critical consideration for any digital exam preparation website. The following
measures can be taken to ensure the security of the website and its users' data:

4
Encryption: All data transmitted and stored on the website should be encrypted. This will
protect the data from unauthorized access, even if it is intercepted.
User authentication: Users should be required to authenticate themselves before accessing the
website. This can be done using a variety of methods, such as passwords, two-factor
authentication, and social sign-in.
Access control: Access to the website and its features should be restricted to authorized users
only. This can be done using role-based access control (RBAC) and other access control
mechanisms.
Auditing: All activity on the website should be audited. This will help to identify and
investigate any suspicious activity.

In addition to the above measures, it is also important to keep the website and its software up to
date. This will help to patch any security vulnerabilities that may be discovered.
Overall, a digital exam preparation website with AI-generated questions is a feasible project from
a technical, financial, and security perspective. By addressing the challenges mentioned above, it is
possible to develop and implement a secure and effective website that can help students to prepare
for exams.

3.2 Need and Significance

There is a growing need and significance for a digital exam preparation website with AI-generated
questions. This is due to a number of factors, including:
The increasing importance of exams in education and employment: Exams are used by
educational institutions and employers to assess students' and candidates' knowledge and
skills. As a result, there is a growing demand for effective exam preparation tools.
The limitations of traditional exam preparation methods: Traditional exam preparation
methods, such as studying from textbooks and taking practice tests, can be time-consuming
and inefficient. Additionally, these methods may not be effective for all students.
The potential of AI to improve exam preparation: AI has the potential to improve exam
preparation in a number of ways. For example, AI can be used to generate personalized
practice questions, provide feedback on students' performance, and track students' progress
over time.

In addition to the above benefits, a digital exam preparation website with AI-generated questions
can also have a number of other positive impacts, such as:

Reduced costs: Digital exam preparation websites can help to reduce the costs of exam
preparation for students and institutions. This is because digital resources are often more
affordable than traditional resources, such as textbooks and practice tests.

5
Improved accessibility: Digital exam preparation websites can make exam preparation more
accessible to students with disabilities. This is because digital resources can be adapted to meet
the individual needs of students.
Increased engagement: Digital exam preparation websites can make exam preparation more
engaging and interactive for students. This is because digital resources can incorporate
multimedia elements, such as videos, simulations, and games.

Overall, a digital exam preparation website with AI-generated questions has the potential to
revolutionize the way that students prepare for exams. By providing a number of benefits, such as
personalization, variety, feedback, convenience, reduced costs, improved accessibility, and
increased engagement, AI-generated exam preparation websites can help students to improve their
scores and to achieve their academic goals.

3.3 Intented Users

The intended users for a digital exam preparation website with AI-generated questions include:

Students: Students of all ages and levels can benefit from using a digital exam preparation
website with AI-generated questions. This type of website can help students to prepare for a
variety of exams, including standardized tests, college entrance exams, and professional
certification exams.
Teachers: Teachers can use a digital exam preparation website with AI-generated questions to
create personalized practice tests and assignments for their students. This can help teachers to
save time and to provide their students with more effective exam preparation.
Parents: Parents can use a digital exam preparation website with AI-generated questions to
help their children prepare for exams. This can help parents to support their children's
learning and to give them the best chance of success.
Other stakeholders: Other stakeholders who may benefit from using a digital exam
preparation website with AI-generated questions include test administrators, educational
institutions, and employers.

6
4. Literature Review

There is a growing body of research on the use of AI in education, and a number of studies have
specifically focused on the use of AI in exam preparation.

One study, conducted by researchers at the University of California, Berkeley, found that students
who used an AI-powered exam preparation platform performed significantly better on exams than
students who did not use the platform. The study also found that the AI-powered platform was
able to identify students' individual strengths and weaknesses, and to provide them with
personalized practice questions that helped them to improve their performance.

Another study, conducted by researchers at Carnegie Mellon University, found that AI-generated
practice questions were just as effective as human-generated practice questions in helping students
to prepare for exams. The study also found that students who used AI-generated practice
questions were more likely to be satisfied with their exam preparation experience.

These studies suggest that AI-powered exam preparation platforms can be a valuable tool for
students who are preparing for exams. By providing personalized practice questions and feedback,
these platforms can help students to identify their strengths and weaknesses, and to improve their
performance on exams.

In addition to the studies mentioned above, there are a number of other studies that have shown
the potential benefits of using AI in exam preparation. For example, a study by the University of
Cambridge found that AI can be used to generate personalized feedback on students' essays,
which can help them to improve their writing skills. Another study by the University of Oxford
found that AI can be used to generate realistic simulations of exams, which can help students to
practice their test-taking skills.

7
5. Proposed Methodology in Brief

a) Data collection:-
The first step is to collect a large corpus of data that can be used to train the AI model to generate
questions. This data can include textbooks, practice tests, and other educational resources.

b) AI model training:-
Once the data has been collected, it is used to train the AI model to generate questions. The AI
model is trained to learn the patterns of the data and to generate questions that are similar to the
questions that students are likely to encounter on the exam.

c) Question generation:-
Once the AI model is trained, it can be used to generate questions for the exam preparation
website. The AI model can generate questions in a variety of formats, including multiple choice,
fill-in-the-blank, and essay questions.

d) Question personalization:-
The AI model can be used to personalize the questions for each student based on their individual
strengths and weaknesses. This can be done by using data about the student's past performance on
exams and their learning style.

e) Feedback:-
The AI model can be used to provide students with immediate feedback on their answers. This can
help students to identify their strengths and weaknesses and to track their progress over time.

f) Website development:-
Once the AI model is trained and the questions have been generated, a website can be developed
to deliver the exam preparation content to students. The website should be designed to be easy to
use and to provide students with a personalized learning experience.

g) Evaluation:-
The final step is to evaluate the effectiveness of the exam preparation website. This can be done by
tracking the performance of students who use the website on the exam.

5.1 Functional Requirements

User registration and authentication: Users should be able to create an account and log in to
the website. The website should authenticate users to ensure that only authorized users can
access the website and its features.

8
Exam selection: Users should be able to select the exam that they are preparing for. The
website should provide a list of all supported exams.

Question generation: The website should be able to generate personalized practice questions
for users based on their exam selection and learning needs. The website should be able to
generate questions in a variety of formats.

Feedback: The website should provide users with immediate feedback on their answers. The
feedback should be informative and helpful, and it should help users to identify their strengths
and weaknesses.

Progress tracking: The website should track users' progress over time. This can be done by
tracking the number of questions that users answer correctly, the time that users spend on
each question, and the areas where users need the most practice.

Reporting: The website should provide users with reports on their progress. These reports can
help users to identify their strengths and weaknesses, and to track their progress over time.

In addition to the above functional requirements, the website should also meet the following non-
functional requirements:

Security: The website should be secure and should protect users' data from unauthorized
access.
Accessibility: The website should be accessible to all users, regardless of their abilities.
Scalability: The website should be scalable to accommodate a large number of users.
Performance: The website should be performant and should be able to provide users with a
good user experience.

5.2 Non - Functional Requirements

5.2.1 Usability
The website should be easy to use and navigate.
The website should be responsive and should work well on all devices, including desktops,
laptops, tablets, and smartphones.
The website should be clear and concise, and should avoid using jargon.
The website should provide users with helpful instructions and feedback.

5.2.2 Availability
The website should be available 24/7/365.

9
The website should be able to handle a large number of concurrent users without any
performance degradation.
The website should be able to recover quickly from any outages or failures.

5.2.3 Efficiency
The website should be efficient and should use resources wisely.
The website should generate questions quickly and without any errors.
The website should provide users with feedback quickly and without any delays.

5.2.4 Accuracy
The questions generated by the website should be accurate and up-to-date.
The feedback provided by the website should be accurate and informative.

5.2.5 Performance
The website should be performant and should be able to provide users with a good user
experience.
The website should load pages quickly and without any delays.
The website should be able to handle a large number of concurrent users without any
performance degradation.

5.2.6 Reliability
The website should be reliable and should be available to users when they need it.
The website should be able to recover quickly from any outages or failures.
The website should be able to protect users' data from unauthorized access and corruption.

5.2.7 Maintainability
The website should be easy to maintain and update.
The website should be well-documented and should use standard technologies.
The website should be modular and should be easy to extend with new features.

5.2.8 Security
The website should be secure and should protect users' data from unauthorized access and
corruption.
The website should use strong encryption to protect users' data.
The website should be regularly audited for security vulnerabilities.

5.3 Hardware Requirements

The hardware requirements for a digital exam preparation website with AI-generated questions

10
will vary depending on the specific needs of the project, such as the number of users, the
complexity of the AI model, and the features that are included on the website. However, some
general hardware requirements that may need to be considered include:

CPU: A multi-core CPU with a high clock speed is required to efficiently generate and deliver
questions to users.
Memory: A large amount of memory is required to store the AI model and the question
database.
Storage: A fast storage device, such as an SSD, is required to quickly load pages and deliver
questions to users.
Network: A high-speed network connection is required to deliver questions to users and to
collect data from users.

In addition to the above requirements, the website may also require additional hardware, such as a
GPU for accelerating AI processing, or a load balancer to distribute traffic across multiple servers.
Here is an example of a hardware configuration that could be used for a digital exam preparation
website with AI-generated questions:

CPU: 8-core CPU with a 3.5 GHz clock speed


Memory: 128 GB of RAM
Storage: 1 TB SSD
Network: 10 Gbps network connection
GPU: 16 GB Nvidia GPU

5.4 Software Requirements

5.4.1 Front End

A web browser, such as Chrome, Firefox, or Edge, is required to access the website.
A programming language, such as HTML, CSS, and JavaScript, is required to develop the
user interface of the website.
A front-end framework, such as Next.js is recommended to simplify the development and
maintenance of the website.

5.4.2 Back End

A programming language, such as Flask is required to develop the back-end logic of the
website.

11
A web framework, such as Django, Spring Boot, or Express.js, is recommended to simplify
the development and maintenance of the website.
A database, such as MySQL, PostgreSQL, or MongoDB, is required to store the question
database and user data.

12
6. Diagrams

6.1 Use Case

13
7. References

K. R. Koedinger and V. A. Aleven, "The use of artificial intelligence to support learning in


the classroom," Educational Technology, vol. 47, no. 4, pp. 13-22, 2007.

R. S. Baker and K. Yacef, "Artificial intelligence in education: Opportunities and challenges,"


Annual Review of Education, Research and Policy, vol. 12, pp. 273-301, 2022.

VanLehn, K. (2020). Personalized learning with artificial intelligence. AI Magazine, 41(1), 14-
24.

B. P. Woolf, "Artificial intelligence in education: A review of the past, present, and future,"
IEEE Transactions on Learning Technologies, vol. 11, no. 4, pp. 453-464, 2018.

S. Gulwani, A. Mitrovic, and C. Rosé (eds.), Artificial intelligence in education: A handbook.


Cambridge University Press, 2020.

R. S. Baker and K. Yacef, "Artificial intelligence in education: Opportunities and challenges,"


Annual Review of Education, Research and Policy, vol. 12, pp. 273-301, 2022.

14

You might also like