SAVITRIBAI PHULE PUNE UNIVERSITY

A PROJECT REPORT ON

Blooms Taxonomy Based Question Paper Generation Using Machine

Learning Algorithms
SUBMITTED TOWARDS THE
PARTIAL FULFILLMENT OF THE REQUIREMENTS OF

BACHELOR OF ENGINEERING (Computer Engineering)

BY
SHRUTI JADHAV (B1904404226)
PIYUSH KALE (B1904404234)
SHUBHAM RAJOLE (B1904404268)
ABHISHEK YADAV (B1904404297)

Under The Guidance of

Prof. Apeksha Pande

DEPARTMENT OF COMPUTER ENGINEERING

Siddhant College of Engineering
Sudumbare, Talegaon Chakan Highway, Pune 410501
Academic year 2024-2025

1
 Siddhant College of Engineering
DEPARTMENT OF COMPUTER ENGINEERING

CERTIFICATE
This is to certify that the Project Entitled
Blooms Taxonomy Based Question Paper Generation Using
Machine Learning AlgorithmsSubmitted by

SHRUTI JADHAV (B1904404226)

PIYUSH KALE (B1904404234)
SHUBHAM RAJOLE (B1904404268)
ABHISHEK YADAV (B1904404297)

is a bonafide students of this institute and the work has been carried out by them under the
supervision of [Link] Pande and it approved for the partial fulfillment of the
requirement of Savitribai Phule Pune University, for the award of degree of Bachelor of
Computer Engineering.

Dr. L V Kamble

Prof. Meenal Wagh Principal

HOD
Dept of Computer Engg SCOE

Prof. Apeksha Pande

Project Co-ordinator External Examiner

2
 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

Acknowledgments

It gives us great pleasure in presenting the preliminary project report on ‘Blooms

Taxonomy Based Question Paper Generation Using Machine Learning Algorithms’.

We would like to take this opportunity to thank our internal guide [Link] Pande and HOD
Prof. Meenal Wagh for giving us all the help and guidance which we needed. We are really
grateful to them for their kind support. Their valuable suggestions were very helpful.

At last we must express our sincere heartfelt gratitude to our Parents who helped us directly
or indirectly during this project work.

SHRUTI JADHAV (B1904404226)

PIYUSH KALE (B1904404234)
SHUBHAM RAJOLE (B1904404268)
ABHISHEK YADAV (B1904404297)
(B.E. Computer Engg.)

S.C.O.E., Department of Computer Engineering Year 2024-25

I
 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

Abstract

In the modern educational landscape, the efficient creation of assessment materials remains a
critical challenge. This project addresses this challenge through the development of a
sophisticated question paper generation system that leverages Bloom's Taxonomy and machine
learning techniques. The system allows faculty members to upload question datasets, from which it
autonomously identifies the cognitive complexity of questions and allocates appropriate marks.
Utilizing machine learning algorithms, such as linear regression for mark prediction and random
forest for Bloom's Taxonomy level prediction, the system streamlines the question paper
generation process, saving time and ensuring accuracy. Through extensive testing and validation,
the effectiveness and reliability of the system have been demonstrated, offering a promising
solution to enhance assessment practices in educational institutions. The key features of the system
include the alignment with Bloom's Taxonomy, ensuring well-rounded assessments, the automated
allocation of marks based on question attributes through machine learning algorithms, reducing
manual effort, and the generation of user-friendly PDF question papers. Thorough testing and
validation procedures have been implemented to ensure the accuracy and reliability of the system's
outputs. This innovative approach to question paper generation has the potential to transform
assessment practices in educational institutions, empowering faculty members to create more
effective and efficient assessments while upholding the principles of Bloom's Taxonomy.

S.C.O.E., Department of Computer Engineering Year 2024-25

II
 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

INDEX
1 Synopsis 1
1.1 Project Title . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2 Project Option . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.3 Internal Guide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.3.1Sponsorship and External Guide . . . . . . . . . . . . . . . . . . . . . . . . 2
1.3.2Technical Keywords (As per ACM Keywords) . . . . . . . . . . . . . . . . . 2
1.4 Problem Statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.5 Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.6 Goals and Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.6.1 Relevant mathematics associated with the Project . . . . . . . . . . . . . . .4
1.6.2 Names of Conferences / Journals where papers can be published . . . . . . .5
1.6.3 Review of Conference/Journal Papers supporting Project idea . . . . . . . . .5
1.7 Plan of Project Execution . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

2 Technical Keywords 9
Area of Project ................................................................................................... 10
Technical Keywords ............................................................................................ 10

3 Introduction 11
3.1 Project Idea .................................................................................................... 12
3.2 Motivation of the Project................................................................................ 15
3.3 Literature Survey ........................................................................................... 16

4 Problem Definition and scope 19

4.1 Problem Statement ......................................................................................... 20
4.1.1 Goals and objectives ....................................................................... 20
4.1.2 Statement of scope ......................................................................... 20
4.2 Methodologies of Problem solving and efficiency issues ................................ 21
4.3 Outcome ........................................................................................................ 22
4.4 Applications .................................................................................................. 22
4.5 Hardware Resources Required ....................................................................... 22

S.C.O.E., Department of Computer Engineering Year 2024-25

V
 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

4.6 Software Resources Required ........................................................................ 23

5 Project Plan 24
5.1 Project Estimates ........................................................................................... 25
5.1.1 Reconciled Estimates...................................................................... 25
5.1.2 Project Resources ........................................................................... 27
5.2 Risk Management w.r.t. NP Hard analysis ..................................................... 27
5.1.2 Risk Identification .......................................................................... 27
5.3 Project Schedule ............................................................................................ 28
5.3.1 Project task set ............................................................................... 28
5.4 Team Organization ........................................................................................ 28
5.4.1 Management reporting and communication .................................... 28

6 Requirement specification relevant to mathematics derived and software engg. 29

6.1 Introduction ................................................................................................... 30
6.1.1 Purpose and Scope of Document .................................................... 30
6.1.2 Overview of responsibilities of Developer ...................................... 31
6.2 Usage Scenario .............................................................................................. 31
6.2.1 Use Case View ............................................................................... 31
6.3 Data Model and Description .......................................................................... 32
6.3.1 Data Description............................................................................. 32
6.4 Functional Model and Description ................................................................. 32
6.4.1 Data Flow Diagram ........................................................................ 32
6.4.2 Activity Diagram: ........................................................................... 35
6.4.3 Functional Requirements ................................................................ 36
6.4.4 Non Functional Requirements: ....................................................... 36
6.4.5 Limitations of Existing System ....................................................... 38
6.4.6 Software Interface Description ....................................................... 38

7 Detailed Design Document 39

7.1 Introduction ................................................................................................... 40
7.2 Architectural Design ...................................................................................... 41
7.3 Data design (using Appendices A and B) ....................................................... 42

S.C.O.E., Department of Computer Engineering Year 2024-25

VI
 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

7.3.1 Internal software data structure ....................................................... 42

7.3.2 Global data structure ...................................................................... 42
7.4 Component Design ........................................................................................ 43
7.4.1 Class Diagram ................................................................................ 43

8 Project Implementation 44
8.1 Introduction ................................................................................................... 45
8.2 Tools and Technologies Used ........................................................................ 45
8.3 Methodologies ............................................................................................... 45

9 Software Testing 47
9.1 Type of Testing Used .................................................................................... 48
9.1.1 Testing Strategy ............................................................................. 48
9.1.2 Testing Levels ................................................................................ 48
9.2 Test Cases Applied for the Testing. ............................................................... 50

10 Results 56
10.1 Screen shots ........................................................................................................ 57

11 Deployment and Maintenance 61

11.1 Installation and un-installation ..................................................................... 62
11.2 User help ..................................................................................................... 63

12 Summary and Conclusion 65

13 REFERENCES 67

78

Annexure Information of Project Group Members 83

S.C.O.E., Department of Computer Engineering Year 2024-25

VII
 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

List of Figures
[Link]. Name of Figure Page No.
6.2.1 Use case Diagram 27
[Link] DFD Level 0 28
[Link] DFD Level 1 28
[Link] DFD Level 2 29
6.4.2 Activity Diagram 29
7.2 Architectural Design 33
7.3.2 E-R Diagram 34
7.4 Component Diagram 34
7.4.1 Class Diagram 35
10.1.1 Dashboard 46
10.1.2 Registration page 47
10.1.3 Login page 48
10.1.4 Template 49
10.1.5 Generate Question Paper 50
10.1.6 Final output 51

8 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

List of Tables
[Link]. Name of Table Page No.
1.6 Plan of project Execution 13
5.1.2 Project Resources 23
5.2.1 Identified Risks and Mitigation 24
5.3.1 Project Task Set 25
5.4.1 Communication Plan 25
6.3.1 Data Description 28
9.2.1 Test cases for user registration module 40
9.2.2 Test cases for user login module 41
9.2.3 Test cases for trained data set module 42
9.2.4 Test cases of blooms level prediction 43
9.2.5 Test cases for marks prediction module 44
9.2.6 Test cases for question paper generation module 45

9 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

CHAPTER 1

SYNOPSIS

1.1 PROJECT TITLE

Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

1.2 PROJECT OPTION

Academic Project

1.3 INTERNAL GUIDE

[Link] Pande

1.3.1 SPONSORSHIP AND EXTERNAL GUIDE

No sponsorship

1.3.2 TECHNICAL KEYWORDS (AS PER ACM KEYWORDS)

Primary Keywords

 Artificial Intelligence (AI) → Machine Learning

 Computing methodologies → Supervised learning
 Computing methodologies → Natural language processing
 Computing methodologies → Feature selection
 Applied computing → Computer-assisted instruction
 Applied computing → E-learning
 Software and its engineering → Software development techniques
 Software and its engineering → Reusable software

Secondary Keywords

 Information systems → Information retrieval

 Information systems → Data mining
 Mathematics of computing → Statistical learning models
 Human-centered computing → Interactive systems
 Human-centered computing → User studies
 Theory of computation → Models of learning
 Computing methodologies → Classification and regression

1.4 PROBLEM STATEMENT

To develop a system that assigns marks to question based on Bloom’s taxonomy and
generates Question papers in orders to reduce human error and effort while consuming
less time.

10 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

1.5 Goals and objectives

The primary goal and objective of this project is to develop an automated system for generating
question papers that integrates machine learning algorithms and Bloom's Taxonomy. The system
aims to enhance the efficiency, accuracy, and quality of the examination process by addressing
several key goals:

Automation of Question Paper Generation

The project seeks to automate the traditionally manual process of question paper creation, reducing
the time and effort required from educators while maintaining high standards of quality and
relevance.

Incorporation of Bloom's Taxonomy for Balanced Assessment

By integrating Bloom's Taxonomy, the system ensures that the generated questions are distributed
across various cognitive levels, from basic knowledge recall to higher-order thinking skills such as
analysis, synthesis, and evaluation.
This allows for a comprehensive and pedagogically sound assessment that aligns with educational
objectives.

Reduction of Human Error

The automated system aims to eliminate common human errors in question paper creation, such as:
Repeated questions, Misalignment with learning outcomes, Inconsistent difficulty levels, This
contributes to more accurate and reliable assessments.

Customizable Question Papers

The project allows educators to tailor question papers to specific requirements, such as:
 Subject matter
 Difficulty level
 Number of questions
This ensures that the exams are adaptable to different curricula and institutional guidelines, making
the system flexible and versatile.

Scalability and Accessibility

The project aims to develop a scalable system capable of generating question papers for a wide
range of educational institutions, from schools to universities.
Additionally, the feature of generating exam papers in PDF format enhances accessibility and ease
of use for educators.

Support for Continuous Improvement and Innovation in Education

By leveraging machine learning and artificial intelligence, the project contributes to ongoing
innovation in education technology (EdTech).
It promotes the use of computational tools in enhancing the teaching and learning process, making
assessments more effective and efficient.

Simplifying the Workflow for Educators

One of the objectives is to streamline the overall examination preparation process, allowing
educators to focus more on:
 Instruction
 Student development
This reduces their workload on administrative tasks related to exam creation.

11 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

1.5.1 Relevant mathematics associated with the Project

The mathematical module plays a pivotal role in ensuring that the question paper generation
process is based on logical and structured algorithms, which adhere to certain educational
frameworks (like Bloom’s Taxonomy). The module incorporates several mathematical models and
operations to ensure efficient, balanced, and randomized selection of questions. And for the
classification is done using two machine learning algorithms i.e linear regression and random
forest.

Linear Regression Algorithm :-

Linear regression aims to model the relationship between a dependent variable ( Y ) and one or
more independent variables ( X ). The equation for a simple linear regression is:
Y=β0+β1X+ϵ
Where:
 ( Y ) is the dependent variable.
 ( X ) is the independent variable.
 ( \beta_0 ) is the intercept.
 ( \beta_1 ) is the slope coefficient.
 ( \epsilon ) is the error term.

For multiple linear regression, the equation extends to:

Y=β0 +β1 X1 +β2 X2 +…+βn Xn +ϵ

The coefficients ( \beta ) are estimated using the least squares method, which minimizes the sum of
the squared residuals:

Application in Bloom’s Taxonomy:

Linear regression can be used to predict the difficulty level of questions based on various features
such as the number of keywords, length of the question, and the cognitive level required (e.g.,
remembering, understanding, applying). This helps in classifying questions into different levels of
Bloom’s Taxonomy.

Random Forest Algorithm

Random Forest is an ensemble learning method that constructs multiple decision trees during
training and outputs the mode of the classes (classification) or mean prediction (regression) of the
individual trees.

The steps are:

1. Bootstrap Sampling: Randomly sample with replacement from the training data.
2. Tree Construction: For each sample, construct a decision tree using a random subset of
features at each split.
3. Aggregation: Aggregate the predictions from all trees.

The prediction for classification is:

y^=mode(y1,y2,…,y T)

12 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

For regression, it is:

Where ( T ) is the number of trees, and ( y_t ) is the prediction from the ( t )-th tree.

Application in Bloom’s Taxonomy:

Random Forest can be used to classify questions into different levels of Bloom’s Taxonomy by
training on labeled data. Features such as question length, complexity, and keywords can be used
to train the model. For question paper generation, the algorithm can ensure a balanced distribution
of questions across different cognitive levels by selecting questions from each category.

Combining Both Algorithms

For a comprehensive system, you can use linear regression to predict the difficulty level of
questions and random forest to classify them into Bloom’s Taxonomy levels. This dual approach
ensures both accurate difficulty prediction and robust classification, aiding in the generation of
well-structured question papers.

1.6 Plan of Project Execution

[Link] Name/Title Start Date End Date Remark

1 Preliminary Survey 05/07/22 04/09/22
2 Literature Survey 04/08/22 17/08/22
3 Introduction and Problem Statement 17/08/22 17/08/22
4 Project Statement 17/09/22 01/09/22
5 Software Requirement and 01/09/22 07/09/22
Specification
6 System Design 07/09/22 15/09/22
7 Partial Report Submission 15/09/22 30/09/22
8 Architecture Design 30/09/22 14/10/22
9 Implementation 14/10/22 02/10/22
10 Deployment 16/12/
11 Testing
12 Paper Publish
13 Report Submission

13 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

Chapter 2

Technical Keywords
2.1 Area of project
The project was implemented in the domain of educational assessment, specifically
focusing on automating the process of question paper generation. The system was designed to cater
to the needs of educators and faculty members across various educational institutions, including
schools, colleges, and universities. By integrating machine learning techniques with Bloom's
Taxonomy, the system aimed to streamline the assessment process and enhance the quality and
effectiveness of educational evaluations. The implementation encompassed developing algorithms
for question categorization, mark allocation, and keyword extraction, as well as designing a user-
friendly web interface for easy interaction

2.2 TECHNICAL KEYWORDS

Machine Learning & AI

 Machine Learning
 Random Forest Algorithm
 Linear Regression Algorithm
 Feature Engineering
 Data Preprocessing
 Model Training & Testing
 Classification Algorithms
 Supervised Learning
 Text Analysis

Bloom’s Taxonomy & Educational Technology

 Bloom's Taxonomy (Remember, Understand, Apply, Analyze, Evaluate, Create)

 Question Classification
 Automated Question Paper Generation
 Cognitive Complexity Analysis
 Educational Assessment
 Exam Paper Generation
 Learning Outcome-Based Assessment

Software Development & Frameworks

 Python Programming
 Django Framework
 Scikit-learn
 Pandas & NumPy
 OpenCV
 Pickle for Model Serialization
 SQLite Database
 REST API for Web Integration

14 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

Data Science & Processing

 Data Cleaning & Transformation

 Dataset Training & Validation
 Big Data Handling
 Statistical Analysis
 Pattern Recognition

System Architecture & Implementation

 Web-Based Application
 Backend Development (Django, Python)
 Frontend Development (HTML, CSS, JavaScript)
 Database Management (SQL/SQLite)
 User Authentication & Security
 PDF Generation for Question Papers
 Cloud Deployment (Optional)

15 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

Chapter 3
Introduction
3.1 Project idea

The Bloom’s Taxonomy-Based Question Paper Generator is an AI-driven system that automates
the creation of structured, balanced, and cognitive-level-based question papers using machine
learning techniques. The project aims to reduce manual effort, ensure fair mark allocation, and
maintain assessment quality in educational institutions.

How It Works:

1. Educators Upload Questions (CSV format).

2. Machine Learning Algorithms Analyze Questions:
1. Random Forest predicts the Bloom’s Taxonomy level (e.g., Remember, Understand,
Apply).
2. Linear Regression predicts mark allocation based on complexity.
3. The System Generates a Well-Balanced Question Paper following educational standards.
4. A PDF Output is Created for easy distribution.

3.2 Motivation of the project

The project is driven by the need for efficient, automated tools to create assessment materials in the
educational sector. This system helps reduce manual work while adhering to Bloom’s Taxonomy,
ensuring a balanced cognitive assessment. The growing need for efficient, accurate, and balanced
assessment systems in educational institutions has driven the demand for automated solutions.
Traditional question paper generation is a time-consuming and subjective task, often lacking
consistent alignment with educational objectives such as Bloom's Taxonomy. The motivation
behind this project stems from the desire to create a system that automates the generation of
question papers, ensuring that they not only adhere to the cognitive levels defined in Bloom's
Taxonomy but also provide accurate and balanced assessments across different difficulty levels.
By leveraging machine learning algorithms, this project aims to reduce the manual effort required
from educators while ensuring fairness, objectivity, and educational rigor in the assessment
process. This automation can save valuable time for educators and enhance the quality of
assessments in both academic and government examination contexts.

3.3 Litetrature Survey

1. Paper Name: Automatic Generation of Question Paper from User Entered Specifications Using
a Semantically Tagged Question Repository

Author: G. Nalawade and R. Ramesh

Abstract: Author There are no standardized methods to ensure quality of question paper. Hence
there arises a need to have a system which will automatically generate the question paper from the
teacher entered specification within a few seconds. Researchers recommend different sets of tags
such as cognitive level, difficulty level, type of question, content /topic for defining a question etc.
The existing tools are rigid and support very basic or limited tags. The proposed system will
automatically generate a question paper from semantically tagged question repository. This system
offers flexibility by supporting all four tags and allows entry of every property in the form of
ranges.

16 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

2. Paper Name: An algorithm for question paper template generation in question paper generation
system.

Author: V. M. Kale and A. W. Kiwelekar

Abstract: In this paper, we present the design of an algorithm to generate question paper template
which satisfies the above-mentioned constraints. The algorithm is illustrated in the paper by using
four constraints namely question paper format, coverage of syllabus, coverage of difficulty levels
and coverage of cognitive levels according to Bloom’s taxonomy. The algorithm presented is
extensible to support any number of user defined constraints.

3. Paper Name: Secure Automatic Question Paper Generation with the Subjective Answer
Evaluation System.

Author: R. Raga Sudha and M. Saravanan

Abstract: This paper describes secure automatic question paper generation and evaluates the
answer which is computerizing in subjective format. For the automatic question paper generation,
the admin must create a database which is the form of question with bloom’s taxonomy. By
clicking the generate button, the question will be generated automatically. To send the question
paper safely to an authorized person cryptography was utilized. For evaluating the answer, admin
must build the question as well as keyword in the database, when the users attend the examination,
it automatically evaluates answer with the keyword matching algorithm. By using this system will
reduce human error and consume less time to generate question and as well as evaluating the
answer.

17 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

Chapter 4
Problem defination and scope
4.1 Problem Statement

To develop a system that assigns marks to question based on Bloom’s taxonomy and
generates Question papers in orders to reduce human error and effort while consuming
less time.

4.1.1 Goals and objective

Automation of Question Paper Generation:
The project seeks to automate the traditionally manual process of question paper creation,
reducing the time and effort required from educators while maintaining high standards of
quality and relevance.

Incorporation of Bloom's Taxonomy for Balanced Assessment:

By integrating Bloom's Taxonomy, the system ensures that the generated questions are
distributed across various cognitive levels, from basic knowledge recall to higher-order
thinking skills such as analysis, synthesis, and evaluation.
This allows for a comprehensive and pedagogically sound assessment that aligns with
educational objectives.

Reduction of Human Error:

The automated system aims to eliminate common human errors in question paper creation,
such as:
Repeated questions, Misalignment with learning outcomes, Inconsistent difficulty levels,
This contributes to more accurate and reliable assessments.

Customizable Question Papers

The project allows educators to tailor question papers to specific requirements, such as:
o Subject matter
o Difficulty level
o Number of questions
This ensures that the exams are adaptable to different curricula and institutional guidelines,
making the system flexible and versatile.

Scalability and Accessibility:

The project aims to develop a scalable system capable of generating question papers
for a wide range of educational institutions, from schools to universities.
Additionally, the feature of generating exam papers in PDF format enhances accessibility
and ease of use for educators.

Simplifying the Workflow for Educators:

One of the objectives is to streamline the overall examination preparation process, allowing
educators to focus more on:
o Instruction
o Student development
This reduces their workload on administrative tasks related to exam creation.

18 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

4.1.2 Statement of scope

Integration of AI for Personalization: Continuing advancements in artificial intelligence

(AI) can further enhance question paper generation systems by enabling personalized
assessments tailored to individual student needs and learning progressions.

 Adaptation to Online and Remote Learning: With the ongoing evolution of online
and remote learning, there's a need for question paper generation systems that
seamlessly integrate with digital platforms and learning management systems. Future
developments could focus on creating adaptable tools that facilitate remote assessment,
feedback, and proctoring mechanisms.
 Enhanced Personalization and Adaptive Learning: Future systems could
incorporate adaptive learning techniques to dynamically adjust the difficulty and
content of assessments based on students' performance and mastery levels. By offering
personalized learning experiences, these systems can better meet the diverse needs and
learning paces of individual students without specifically mentioning machine
learning.
 Multimodal Assessment Formats: Advancements in multimedia technologies offer
opportunities to support the creation of multimodal assessment formats within question
paper generation systems. These formats could include interactive simulations, audio-
visual questions, and virtual reality-based assessments, enhancing student engagement
and learning experiences.

4.2 Methodologies of Problem solving and efficiency issues

1. Methodologies for Problem Solving

To develop an efficient Bloom’s Taxonomy-Based Question Paper Generation System,

the following methodologies were employed:
A. Machine Learning-Based Question Classification
Problem: Traditional question paper generation lacks structured classification, leading to
imbalanced cognitive-level distribution.
Solution:
Random Forest Classifier is used to categorize questions into Bloom’s Taxonomy levels (
Remember, Understand, Apply, Analyze, Evaluate, Create).
This ensures well-balanced assessments covering all cognitive levels.

B. Automated Mark Allocation Using Regression Models

Problem: Manual mark distribution is subjective and inconsistent.
Solution:
Linear Regression Model predicts marks based on question attributes (difficulty, length,
cognitive level).
This eliminates human bias and ensures fair marking.

C. Customizable Question Selection & Filtering

Problem: Teachers need flexibility in difficulty levels, topics, and question selection.
Solution:
The system provides filters for question selection based on subject, topic, difficulty, and

19 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

number of questions.
Users can modify parameters dynamically to generate customized papers.

D. Web-Based System for Accessibility

Problem: Manual question paper generation requires time-consuming paperwork and lacks
accessibility.
Solution:
Developed a user-friendly web application using Django, HTML, CSS, and JavaScript.
Secure authentication for educators, enabling anytime, anywhere access.

E. PDF Generation for Easy Distribution

Problem: Educators manually format question papers, which is tedious and error-prone.
Solution:
The system automatically generates structured question papers in PDF format, reducing
manual effort.

2. Efficiency Issues & Their Solutions

A. Computational Complexity of ML Models
Issue: Training ML models on large datasets can be computationally expensive.
Solution:
Optimized dataset preprocessing using feature selection and dimensionality reduction.
Used efficient libraries (Scikit-learn, NumPy, Pandas) to speed up training.

B. Data Quality & Cleaning Challenges

Issue: Poorly formatted or incorrect data may lead to incorrect classification.
Solution:
Data preprocessing (removal of duplicates, missing values handling, tokenization).
Keyword extraction ensures questions align correctly with Bloom’s Taxonomy.

C. Load Balancing & Scalability

Issue: If multiple users generate papers simultaneously, system performance may degrade.
Solution:
Implemented asynchronous processing for ML model execution.
Used database indexing to speed up question retrieval.

D. Ensuring Question Diversity

Issue: The system may select repeated or overly similar questions.
Solution:
Applied randomization algorithms to ensure variety in selected questions.
Set rules to prevent duplicate questions within a single exam.

E. Security & Authentication

Issue: Unauthorized users may manipulate question papers.
Solution:
Implemented role-based authentication (educators, admins).
Used secure login mechanisms with hashed passwords.

20 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

4.4 Applications

The utilization of a question paper generation system based on Bloom's taxonomy offers numerous
applications within educational contexts:
1. Customized Assessment Development: Educators can leverage the system to craft
assessments tailored to specific learning objectives and cognitive levels outlined by Bloom's
taxonomy. This ensures the alignment of assessment items with desired educational outcomes,
fostering a comprehensive evaluation of students' understanding and critical thinking abilities.
2. Support for Formative Assessment Practices: The system aids in the creation of formative
assessments that allow educators to gauge student progress over time. By generating questions
targeting different cognitive levels, instructors can gain insights into students' learning
processes and areas requiring further attention or reinforcement.
3. Facilitation of Differentiated Instruction: Through the incorporation of varied cognitive-
level questions, the system supports the creation of assessments that cater to diverse learning
needs and abilities. This enables educators to accommodate different learning styles and
challenge students at appropriate skill levels.
4. Enhancement of Curriculum Planning: Administrators and curriculum developers can
utilize the system to ensure coherence between instructional objectives, learning activities, and
assessment strategies. By integrating Bloom's taxonomy into assessment design, educational
stakeholders can uphold consistency in evaluating the attainment of learning goals.
5. Professional Development in Teaching Practices: The system serves as a valuable tool in
teacher training programs, exemplifying effective assessment methodologies and the
application of Bloom's taxonomy in designing meaningful learning assessments. Educators can
learn to create assessments that foster higher-order thinking skills and deepen student
comprehension.
6. Preparation for Standardized Testing: Educational institutions preparing students for
standardized tests can benefit from the system by generating

4.5 Hardware Resources Required

a) Project Development
For project development, a standard desktop or laptop computer is sufficient. The hardware
requirements are minimal, including.
i. Processor: Any modern processor (e.g., Intel Core i3 or equivalent)
ii. RAM: At least 4GB RAM for smooth development experience
iii. Storage: Sufficient storage space for project files and development tools (e.g., IDEs,
libraries)
iv. Operating System: Windows, macOS, or Linux operating system

b) Project Operations
For using the project, users need access to devices with web browsing capabilities, such as.
1. Desktop and Laptop Computers: Standard desktop or laptop computers running
Windows, macOS, or Linux operating systems.
2. Mobile Devices: Smartphones or tablets with web browsers, running iOS or Android
operating systems.

21 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

3. Internet Connection: A stable internet connection is required to access the web-

based interface of the project.

4.6 Software Resources Required

a) Project Development
For project development, the following software tools and technologies are required.
• Programming Languages: HTML, CSS, JavaScript, Python
• Integrated Development Environment (IDE): Visual Studio Code, PyCharm, Sublime
Text, or any preferred text editor
• Version Control: Git for version control and collaboration
• Web Framework: Django for backend development (if applicable)
• Libraries: NumPy, Pandas, scikit-learn for machine learning implementation
• Web Browser: Any modern web browser for testing and debugging.

b) Project Operations
• For using the project, users only need a web browser to access the web-based interface. The
following software components are required.
• Web Browser: Compatible with standard web browsers such as Chrome, Firefox, Safari,
or Edge.
• Operating System: The project's web interface is platform-independent and can be
accessed from an

22 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

Chapter 5
Project Plan
5.1 Project Estimates

5.1.1 Reconciled Estimates

• To estimate the time, cost, and effort required for the project, we use the COCOMO
(Constructive Cost Model) approach:

Effort Estimation Formula:

E=a×(KLOC)bE = a \times (KLOC)^bE=a×(KLOC)b

where:

o E = Effort in Person-Months
o KLOC = Thousands of Lines of Code
o a, b = Project-Specific Constants

Estimated Metrics:

o Approximate 10-15 KLOC (Thousand Lines of Code)

o Effort Required: ~6 Person-Months
o Duration: ~14 Weeks
o Estimated Cost: ₹25,000 – ₹30,000

5.1.2 Project Resources

A. Human Resources

Role Responsibilities Number of Members

Project Manager Planning, execution, risk management 1
Machine Learning
Model training & integration 1
Engineer
Backend Developer API development, database setup 1
Frontend Developer UI/UX design, web interface 1
Tester Software testing & debugging 1

23 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

B. Hardware & Software Resources

Resource Specification
Processor Intel i5/i7 or equivalent
RAM Minimum 8GB
Storage Minimum 256GB SSD
Operating System Windows/Linux/Mac
Software Tools Python, Django, Scikit-learn, SQLite, Pandas, NumPy
Database SQLite or MySQL

5..2 Risk Management with Respect to NP-Hard Analysis

5.2.1 Risk Identification

Complexity Analysis:

 The question selection and optimization problem can be NP-Hard, meaning there is no
polynomial-time solution for the best selection of questions to fit cognitive levels and
difficulty balance.
 ML Model Training Complexity: Training time increases with large datasets, which may
lead to performance bottlenecks.

Identified Risks & Mitigation:

Risk Impact Mitigation Strategy

NP-Hard Problem Slow execution of the question Implement heuristic or greedy algorithms to
Complexity selection algorithm approximate optimal question sets
ML Model Use cross-validation & prune unnecessary
Poor generalization to new data
Overfitting features
System Downtime Affects performance Implement load balancing & caching
Unauthorized access to
Data Security Issues Use role-based authentication & encryption
question data
Duplicate Question Implement randomization & duplicate
Poor question diversity
Selection filtering

24 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

5.3 Project Schedule

5.3.1 Project task set

Phase Tasks Timeline

Phase 1: Requirement Analysis Define scope, objectives, & feasibility study Week 1-2
Phase 2: System Design Design architecture, ER diagrams, database schema Week 3-4
Phase 3: Backend & Machine Develop ML models (Random Forest, Linear
Week 5-7
Learning Development Regression), API integration
Week 8-
Phase 4: Frontend Development UI/UX design, question upload, integration
10
Week
Phase 5: Testing & Debugging Unit testing, UAT, bug fixing
11-12
Phase 6: Deployment & Week 13-
Final deployment, user training, reporting
Documentation 14

5.4 Team Organization

5.4.1 Management reporting and communication

Communication Plan:

Communication Type Frequency Stakeholders

Team Meetings Weekly Development Team
Progress Reports Bi-Weekly Project Manager & Stakeholders
Testing Reports After Testing Phase QA Team & Developers
Client Demonstration Final Week Educators & Institutional Heads

25 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

Chapter 6

Requirement specification relevant to mathematics

derived and software engineering
6.1 Introduction
This section provides an overview of the purpose, scope, and responsibilities involved in the
development of the Bloom’s Taxonomy-Based Question Paper Generation System.

6.1.1 Purpose and Scope of Document

Purpose: This document serves as a technical reference for the development, implementation, and
deployment of the automated question paper generation system. It defines:

 The functional and non-functional requirements.

 The responsibilities of developers.
 The usage scenarios and data models.

Scope: The project aims to automate the process of generating question papers by using machine
learning algorithms to classify questions based on Bloom’s Taxonomy and allocate marks
intelligently. The system allows customization for educators and provides PDF output for easy
distribution.

6.1.2 Overview of responsibilities of Developer

Developer Responsibilities:
1. Machine Learning Engineer
a) Train Random Forest Model for question classification.
b) Train Linear Regression Model for automated mark allocation.
c) Implement model validation & optimization.

2. Backend Developer
a) Develop Django-based API for data storage & retrieval.
b) Implement database models for questions, users, and settings.
c) Ensure secure authentication & role-based access control.

3. Frontend Developer
a) Design user-friendly UI using HTML, CSS, JavaScript.
b) Integrate ML models with frontend for seamless interaction.
c) Implement customization options for educators.

4. QA Tester
a) Conduct unit, integration, and system testing.
b) Identify and resolve bugs and performance issues.
c) Validate ML model accuracy and question paper correctness.

26 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

6.2 Usage Scenario

6.2.1 Use Case View

Primary Actors:
ii. Educators/Teachers: Upload questions, customize paper, generate PDFs.
iii. Admin: Manages users, settings, and system configurations.
iv. Machine Learning Model: Classifies questions and assigns marks.

Figure 6.2.1 : Usecase diagram

27 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

6.3 Data Model and Description

6.3.1 Data Description

A. Database Schema

Table Name Description

Users Stores educator/admin details
Questions Contains questions with attributes (taxonomy level, marks, topic)
Generated_Papers Stores generated question papers
ML_Models Stores trained ML models for classification

B. Question Data Format (Example Entry in CSV)

Question Cognitive Predicted

Question Text Difficulty Topic
ID Level Marks
Computer
101 What is AI? Easy Remembering 5
Science
Explain Neural Machine
102 Hard Analyzing 10
Networks. Learning

6.4 Functional Model and Description

6.4.1 Data Flow Diagram

Figure [Link] : Data Flow (0) diagram

Figure [Link] : Data Flow (1) diagram

28 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

Figure [Link] : Data Flow (2) diagram

6.4.2 Activity Diagram:

Figure 6.4.2 : Activity Diagram

29 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

6.4.3 Functional Requirements

1. User Authentication: The system shall provide secure user authentication mechanisms for
administrators and educators to access the system.
2. Question Input: Users shall be able to input questions, including the question content, type,
and associated metadata such as subject, topic, and difficulty level.
3. Bloom's Taxonomy Integration: The system shall categorize questions based on Bloom's
Taxonomy to align with cognitive levels such as remembering, understanding, applying,
analyzing, evaluating, and creating.
4. Machine Learning Algorithms: The system shall implement machine learning algorithms,
including Random Forest and Linear Regression, for tasks such as question categorization and
complexity assessment.
5. Question Customization: Educators shall have the ability to customize question parameters,
including type, difficulty level, and topic coverage, according to their preferences.
6. PDF Generation: The system shall generate question papers in PDF format for easy
distribution to examiners and students.
7. Analytics Dashboard: The system shall provide an analytics dashboard for administrators and
educators to track and analyze student performance on assessments, including question-level
insights and assessment outcomes.

6.4.4 Non Functional Requirements:

1) Performance: The system shall respond swiftly to user requests for question paper generation,
ensuring minimal latency and downtime.
2) Security: The system shall implement robust user authentication and data protection measures
to ensure the confidentiality and integrity of user data.
3) Usability: The user interface shall be intuitive and user-friendly, with clear navigation and
informative feedback messages.
4) Compatibility: The system shall be compatible with standard web browsers such as Chrome,
Firefox, Safari, and Edge, ensuring accessibility across different platforms and devices.
5) Constraints: The system's performance may be affected by the size of the question database
and the complexity of machine learning algorithms.
6) Assumptions: It is assumed that users have basic knowledge of web browsing and computer
usage.
7) Dependencies: The system's functionality may depend on external libraries and frameworks
such as scikit-learn, Django, and NumPy.

6.4.5 Limitations of Existing System

Traditional manual question paper generation and existing automated systems have several
limitations that impact efficiency, accuracy, and scalability. Below are the key limitations of the
current systems:

1. Time-Consuming Process

 Manual selection, categorization, and formatting of questions take significant time for
educators.
 Sorting questions based on Bloom’s Taxonomy is labor-intensive.

2. Lack of Standardization & Balance

 Existing manual question papers often lack balanced difficulty levels.

30 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

 No consistency in ensuring proper distribution of cognitive levels (Remember, Understand,

Apply, etc.).

3. Prone to Human Errors

 Repeated questions, incorrect mark allocation, and misaligned learning outcomes frequently
occur.
 Educators manually adjust question papers, increasing the risk of mistakes.

4. No Automation for Mark Allocation

 In traditional systems, marks are assigned manually, which can be subjective and
inconsistent.
 No intelligent system exists to automatically allocate marks based on question complexity.

5. Limited Customization & Scalability

 Existing systems do not provide dynamic customization options for:

o Selecting questions based on difficulty level.

o Filtering questions by topic or subject.
o Adjusting the number of questions per section.
 Not scalable for institutions handling large question banks.

6. No Integration of Machine Learning

 Traditional question paper generation does not leverage AI/ML for:

o Automatic categorization of questions.

o Predictive mark allocation.
o Pattern recognition to avoid duplicate questions.

7. Poor Security & Accessibility

 Manual paper creation lacks data security, making question leakage a concern.
 No role-based access, meaning unauthorized users could manipulate questions.
 No centralized cloud storage, making it hard to access and share question papers.

8. Inefficient Paper Formatting & Output

 Educators manually format question papers, leading to inconsistent layouts.

 No automated PDF generation, making paper distribution cumbersome.

6.4.6 Software Interface Description

The Bloom’s Taxonomy-Based Question Paper Generation System is a web-based application

designed for educators and administrators to generate structured question papers using machine
learning techniques. This section describes the user interfaces, software interactions, and system
components.

1. User Interface (UI) Description

1.1 Login & Authentication Page

31 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

Purpose: Secure authentication for educators & admins.

Features:

 User Login (Email & Password).

 Registration for new users.
 Role-based access control (Admin/Teacher).
 Forgot Password option.

1.2 Dashboard
purpose: Main control panel for educators.
Feature:

 Overview of uploaded questions & previous question papers.

 Quick access to question upload & paper generation.
 Download & view past generated question papers.

1.3 Question Upload Interface

Purpose: Allow users to upload question datasets (CSV format).
Features:

 Upload CSV file containing questions, topics, difficulty levels.

 System automatically classifies questions based on Bloom’s Taxonomy.
 Display uploaded questions in a structured format.

1.4 Question Paper Customization Panel

Purpose: Provide educators with filters & customization options.
Features:

 Select subject, difficulty level, cognitive level (Bloom's Taxonomy), and number of
questions.
 Option to manually select questions or allow AI-based auto-selection.
 Preview final selection before generation.

1.5 Automated Question Paper Generation

Purpose: Generate a structured & balanced question paper.
Features:

 AI-driven mark allocation based on Linear Regression.

 Categorization using Random Forest (Bloom’s Taxonomy levels).
 Ensure no duplicate questions in the generated paper.
 Automatically generate a PDF file of the question paper.

1.6 PDF Export & Download Page

Purpose: Provide educators with printable question papers.

Features:

 Download generated question papers as PDFs.

 Export papers with customized formatting & institutional branding.
 Maintain a history of generated question papers for future reference.

32 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

Chapter 7

Detailed Design Document

7.1 Introduction
The Detailed Design Document provides a comprehensive overview of the system’s
architecture, components, and interactions. It serves as a blueprint for developers to
implement the Bloom’s Taxonomy-Based Question Paper Generation System efficiently.

7.2 Architectural Design

Figure 7.2 : Activity diagram

7.3 Data design (using Appendices A and B)

The data design includes structured storage mechanisms and relationships between various entities in
the system.

7.3.1 Internal software data structure

The system uses structured data formats to ensure efficient storage and retrieval. Key internal
structures include:
 Question Bank: Stores questions along with attributes like difficulty level, topic, and
Bloom’s taxonomy classification.
 User Data: Maintains user credentials and roles.
 Generated Question Papers: Keeps track of generated exams for reuse or modification.

7.3.2 Global data structure

The Entity-Relationship (E-R) diagram in Fig 4.2 illustrates a system designed for managing exam
papers, likely tailored for a university or college setting. Entities include Branch representing
academic departments, Examiner for individuals creating exam papers, and Question Paper denoting
specific exam papers. Additionally, relationships like Prepares linking Examiner to Question Paper
and Contains connecting Set of Questions to individual Questions are depicted. Although
cardinalities are absent, the diagram effectively portrays the core entities and their associations within
the exam paper management system, facilitating efficient organization and retrieval of assessment
materials.

33 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

Fig 7.3.2 : E-R Diagram

7.4 Component Design
The diagram depicts a two-part process within the Blooms Taxonomy Based Question Paper Generation
system, where users interact with the system by adding questions to a question bank. These questions are
utilized to train two models: a Linear Regression Model, predicting future assessment marks based on
assessment features, and a Random Forest Model, forecasting Bloom's levels (question complexity) of
future assessments. Subsequently, the system generates question papers based on learning objectives or
student ability levels, leveraging insights gleaned from the trained models to enhance assessment
accuracy and effectiveness.

Fig 7.4 : Component Diagram

34 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

7.4.1 Class Diagram

Fig 7.4.1 : Class Diagram

35 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

Chapter 8
Project Implementation
8.1 Introduction

This section describes the practical implementation of the Blooms Taxonomy-Based Question
Paper Generation System. It details the technologies used and the methodologies followed to
develop and deploy the system efficiently.

8.2 Tools and Technologies Used

 The system employs a variety of tools and technologies to facilitate efficient question paper
generation:
 Programming Languages: Python (Backend), HTML, CSS, JavaScript (Frontend)
 Frameworks: Django (Backend), Bootstrap (Frontend UI Styling)
 Database Management System: SQLite for structured data storage
 Machine Learning Libraries: scikit-learn (Random Forest, Linear Regression), Pandas for
data handling
 Development Tools: Visual Studio Code, GitHub for version control

8.3 Methodologies

Module 1: User Registration

• Module Title: User Registration
• Purpose of Module: Facilitates the registration process for new users by collecting their
information, performing validations, and creating user accounts.
• Inputs and Outputs for Module: Inputs include user information (e.g., name, email,
password), and the output is a user account in the system.
• Files Used by Module: User database for authentication.
• Algorithm: The module collects user information, validates it, and creates a new user
account in the system.
• User Input Collection: Prompt users to enter registration information: name, email, and
password.
• Data Validation: Validate entered data to ensure it meets required criteria (e.g., correct
email format, strong password).
• User Account Creation: If data is valid, create a new user account and store registration
details in the user database.
• Feedback to User: Provide feedback on registration success or errors encountered during
the process.
• Error Handling: Handle any errors gracefully, guiding users to correct issues and resubmit
if necessary.

Module 2: User Login

• Module Title: User Login
• Purpose of Module: Handles the authentication process for users, allowing them to log in to
the system and access other features.
• Inputs and Outputs for Module: Inputs include user login credentials (e.g., email and
password), and the output is a successful login session.
• Files Used by Module: User database for authentication
• Algorithm:-The module verifies user credentials, validates the password, and manages user

36 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

sessions.
• User Input Collection: Prompt users to enter their email and password for login.
• Verification of User Credentials: Retrieve and verify the user's stored information from the
database.
• User Session Management: If credentials are verified, create a session for the user to
maintain their logged-in status.
• Feedback to User: Provide feedback on the success or failure of the login attempt.
• Error Handling: Handle any errors encountered during the login process and guide users
accordingly.
Module 3: Data Set Trained
• Module Title : Data Set Trained
• Purpose of Module: Manages the dataset used for training machine learning models. This
includes importing, preprocessing, and storing the dataset.
• Inputs and Outputs for Module: Inputs include the dataset, and the output is the prepared
dataset for training.
• Files Used by Module: The module utilizes dataset files.
• Algorithm: It involves importing and preprocessing the dataset, ensuring data quality, and
making it ready for training.
• Dataset Importing: Import the dataset containing questions and associated attributes, such
as content and difficulty levels.
• Data Preprocessing: Preprocess the dataset to ensure data quality and compatibility with
machine learning algorithms. Handle missing values, outliers, and any other data
anomalies.
• Feature Engineering: -Extract relevant features from the dataset that can be used to train the
machine learning model. Transform the data into a suitable format for model training.
• Model Training: Utilize machine learning algorithms such as linear regression to train the
model on the prepared dataset. Train the model to identify the complexity of questions and
allocate them marks between 4 to 8 based on their attributes.
• Model Evaluation: Evaluate the trained model's performance using appropriate metrics to
ensure its accuracy and effectiveness in identifying question complexity.
• Model Deployment: Deploy the trained machine learning model to be integrated into the
system for real-time usage in question paper generation.

Module 4: Bloom’s Taxonomy Prediction

• Module Title: Bloom’s Taxonomy Prediction
• Purpose of Module: Predicts the Bloom’s taxonomy level for new questions using machine
learning algorithms, such as Random Forest.
• Inputs and Outputs for Module: Inputs are the question content, and the output is the
assigned Bloom’s taxonomy level.
• Files Used by Module: No specific files are used in this module.
• Algorithm: The module uses machine learning models to predict the Bloom's taxonomy
level based on the question content.
• Input Collection: Receive the content of the question as input.
• Keyword Extraction: Keywords were extracted from questions using a custom algorithm
developed for the purpose. This algorithm effectively identified relevant terms within the
question content that could potentially indicate the cognitive level associated with the
question.
• Alignment with Bloom's Taxonomy: Align the extracted keywords with Bloom's
Taxonomy levels using the custom algorithm. Determine the most appropriate Bloom's
level based on the identified keywords and their relevance to cognitive processes.

37 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

• Prediction: Utilize machine learning algorithms, such as Random Forest, to predict the
Bloom's taxonomy level for the question. Train the model on a labelled dataset of questions
and their corresponding Bloom's levels to enable accurate predictions.
• Output: Output the assigned Bloom's taxonomy level for the question.

Module 5: Mark Prediction

• Module Title: Mark Prediction
• Purpose of Module: Predicts the marks for each question using machine learning
algorithms, such as Linear Regression.
• Inputs and Outputs for Module: Inputs include the question content, Bloom’s taxonomy
level, and other features. The output is the estimated marks.
• Files Used by Module: No specific files are used in this module.
• Algorithm: The module utilizes machine learning models to estimate the marks based on
various factors
• Input Collection: Collect question content, Bloom’s taxonomy level, and other relevant
features as input.
• Feature Engineering: Process input data and engineer additional features that may influence
the marks assigned to each question.
• Model Training: Utilize machine learning algorithms, such as Linear Regression, to train a
model on the prepared dataset.
• Mark Prediction: Use the trained model to predict the marks for each question based on the
input features.
• Output: Output the estimated marks for each question.

Module 6: Question Paper Generation

• Module Title: Question Paper Generation
• Purpose of Module: Generates question papers based on specified criteria, such as the
number of questions and distribution of Bloom’s taxonomy levels.
• Inputs and Outputs for Module: Inputs include criteria for the question paper, and the
output is a well-balanced question paper.
• Files Used by Module: The module may use a question database.
• Algorithm: It selects appropriate questions from the question database, considering the
criteria, to create the final question paper.
• Dashboard Page: Provides a welcoming interface for new users to either log in or register
for access to the system's features.
• Registration Process: Users who do not have an account can navigate to the login page and
click on the "Register" option to sign up for access to the system.
• Admin Access Page: Accessible by administrators, this page allows management of user
accounts and access to the database storing information of signed-up staff.
• Login Page: Displays a login form where existing users can enter their username and
password to access the system.
• Main Page for Question Paper Generation: After successful login, users are redirected to
the main page where they can .Enter the name of the faculty, Select the exam date, Choose
the semester and year and other data that will be reflected in the question [Link] a
CSV file containing [Link] submission, the system validates the details and
generates a question paper using a trained machine learning model to identify question
complexity and Bloom's taxonomy levels.

38 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

Chapter 9
Software Testing
9.1 Type of Testing Used

• The testing phase ensures that the system functions correctly and meets user requirements.
Several testing types were employed:
• Unit Testing: Tests individual components and functions for expected behavior.
• Integration Testing: Ensures that different system modules interact correctly.
• System Testing: Validates the entire system’s functionality against specified requirements.
• User Acceptance Testing (UAT): Conducted with end users to ensure usability and
effectiveness.

9.1.1 Testing Strategy

• A black-box testing approach was used, focusing on inputs and expected outputs rather
than internal code structure. The strategy involved:
• Writing test cases for all major functionalities.
• Automating test execution where possible.
• Conducting performance and stress testing to evaluate system stability.
• Gathering feedback through UAT for refinements.

9.1.2 Testing Levels

• Testing was performed at various levels to ensure reliability:

• Unit Testing: Conducted on individual functions like question categorization.
• Integration Testing: Verified interactions between modules (e.g., database connectivity
with ML models).
• System Testing: Checked complete workflows such as question paper generation.
• Acceptance Testing: Ensured that generated question papers met educational requirements.

39 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

9.2 Test Cases Applied for the Testing.

1. Test Cases of User Registration Module: The test cases for the User Registration Module
ensure that the registration process functions correctly and securely. They validate the user input
validation, error handling, and database integration, guaranteeing a smooth and reliable registration
experience.

Table No. 6.1 Test Cases of User Registration

Test Case ID Description Steps Expected Result Actual Result Status

User will be
successfully
User was successfully
1. Enter valid registered,
registered, redirected
Successful details in all fields. redirected to
TC_001 to confirmation page, Pass
Registration 2. Click Register confirmation page,
and a success message
button. and a success
was displayed.
message will be
displayed.

1. Enter invalid
Error message will
Invalid email address (e.g., Error message
be displayed
TC_002 Email missing @ symbol). "Invalid email format" Pass
indicating invalid
Address 2. Click Register was displayed.
email format.
button.

Error message
Error message will
"Password must be at
1. Enter a weak be displayed
least 8 characters and
password (e.g., indicating at least
Weak include at least one
TC_003 password123). 2. one password Pass
Password uppercase letter, one
Click Register strength
lowercase letter, and
button. requirement not
one number" was
met.
displayed.

1. Enter different
Error message will
passwords in
be displayed Error message
Password Password and Re-
TC_004 indicating "Passwords do not Pass
Mismatch enter Password
passwords do not match" was displayed.
fields. 2. Click
match.
Register button.

40 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

1. Enter an already Error message will Error message "Email

Existing
registered email be displayed address is already
TC_005 Email Pass
address. 2. Click indicating email is registered" was
Address
Register button. already registered. displayed.

2. Test Cases of User Login Module: The test cases for the User Login Module focus on
verifying the functionality and security of the login process. They include scenarios to validate the
correct handling of valid and invalid credentials, authentication mechanisms, session management,
and password encryption.

Table Test Cases of User Login Module

Test
Description Steps Expected Result Actual Result Status
Case ID

1. Enter valid
User successfully User successfully
TC_001 Successful Login credentials. 2. Click Pass
logged in. logged in.
Login.

1. Enter non-existent
Error message Error message
TC_002 Invalid Username username. 2. Click Pass
displayed. displayed.
Login.

1. Enter incorrect
Error message Error message
TC_003 Invalid Password password. 2. Click Pass
displayed. displayed.
Login.

Blank Username 1. Leave both fields Error message Error message

TC_004 Pass
and Password blank. 2. Click Login. displayed. displayed.

1. Leave username Error message Error message

TC_005 Blank Username Pass
blank. 2. Click Login. displayed. displayed.

41 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

3. Test Case of Trained Data Set Module: The test cases for the Trained Data Set Module aim to
verify the integrity and accuracy of the trained data set used in the system. These tests involve
validating the completeness and correctness of the data, ensuring proper data preprocessing techniques
are applied, and confirming that the data is representative of the intended use case

Table Test Case of Trained Data Set Module

Test
Description Steps Expected Result Actual Result Status
Case ID

The module The module

1. Provide a valid dataset successfully successfully
Valid Dataset
TC_001 file (e.g., CSV, Excel) as imports the dataset imported the Pass
Import
input. into a pandas data dataset into a
frame. pandas data frame.

The module threw

1. Provide a file with an Error message
Invalid File an error for the
TC_002 unsupported format (e.g., indicating invalid Pass
Format unsupported file
text file). file format.
type.

Error message The module threw

Missing 1. Don’t provide any
TC_003 indicating missing an error for the Pass
Dataset File input file.
dataset file. missing input file.

The module
Warning message
imported the
Empty Dataset 1. Provide an empty indicating an
TC_004 empty data frame Pass
File dataset file. empty dataset and
but warned the
potential issues.
user.

1. Provide a dataset with The module

The module
missing values. 2. successfully
Basic Data identified and
(Optional) Configure the identifies and
TC_005 Cleaning (if handled the Pass
module to handle missing handles missing
applicable) missing values as
values (e.g., removal or values as
configured.
imputation). configured.

1. Provide a dataset with

The module
mixed data types (e.g., The module
Data Type successfully
strings in numeric successfully
TC_006 Conversion (if converted the data Pass
columns). 2. (Optional) converts data types
applicable) types as
Configure the module to as configured.
configured.
convert data types.

42 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

4. Test Cases of Blooms level prediction: The test cases for Bloom's Level Prediction focus on
evaluating the accuracy and effectiveness of the system in predicting the Bloom's level (question
complexity) of assessments. These test cases involve providing input questions of known Bloom's
levels and comparing the predicted levels generated by the system against the expected values
Table Test Cases of Blooms level prediction

Test
Description Steps Expected Result Actual Result Status
Case ID

The predicted level

Knowledge Provide a question The module must matches the
TC_001 (Remembering) - that tests basic recall predict the Bloom’s question’s Pass
Basic Recall of facts taxonomy level cognitive demand
(recalling a fact).

The predicted level

Provide a question
The module will matches the
Comprehension that requires
predict the Bloom’s question’s
TC_002 (Understanding) - understanding the Pass
taxonomy level as cognitive demand
Grasping Meaning meaning of
“Comprehension”. (understanding a
information
concept).

The predicted level

matches the
Application Provide a question The module will
question’s
(Applying) - Using that requires predict the Bloom’s
TC_003 cognitive demand Pass
Knowledge in applying knowledge taxonomy level as
(applying
New Situations to solve a problem “Application”.
knowledge to a
scenario).

The predicted level

Provide a question
Analysis The module will matches the
that requires
(Analyzing) - predict the Bloom’s question’s
TC_004 analyzing Pass
Breaking Down taxonomy level as cognitive demand
information to
Information “Analysis”. (analyzing
identify relationships
relationships).

Provide a question
that requires new The predicted level
Synthesis ideas or solutions The module will matches the
(Creating) - (e.g., “Design an predict the Bloom’s question’s
TC_005 Pass
Generating New experiment to test the taxonomy level as cognitive demand
Ideas effect of light “Synthesis”. (creating a new
intensity on plant solution).
growth”).

43 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

5. Test Cases for Marks Prediction Module: The test cases for the Marks Prediction Module focus
on verifying the system's ability to accurately predict the marks students will receive on future
assessments. This includes assessing the model's performance with various types of input data,
evaluating the accuracy of predictions across different difficulty levels, and validating the consistency
of results through repeated testing with different datasets

Table Test Cases for Marks Prediction Module

Test
Description Steps Expected Result Actual Result Status
Case ID

Predicted marks
The module will
1. Generate a align with the
Basic predict marks within a
TC_001 question with a set expected range for Pass
Prediction reasonable range for
difficulty level. the chosen
the difficulty level.
difficulty.

The module will Prediction for

1. Generate two
predict slightly higher shorter questions is
open-ended questions
Short vs. Long marks for the shorter higher than, or
TC_002 of similar difficulty Pass
Question question, assuming equal to, the longer
but with varying
similar cognitive question’s
lengths.
processing is required. prediction.

1. Generate two Prediction for the

The module will
open-ended questions question requiring a
Open-Ended predict higher marks
of the same difficulty more complex
TC_003 Question for the question Pass
level but varying in answer is higher
Complexity requiring a more
the complexity of the than the simpler
complex answer.
answer. question.

Prediction deviates
1. Generate a
The module may from labeled
question labeled as
Outlier predict marks higher difficulty but
TC_004 easy but containing a Pass
Difficulty than expected for reflects the actual
complex concept or
“easy” difficulty. complexity of the
logic.
question.

Prediction remains
The module’s
1. Generate a within a reasonable
Random prediction should be
TC_005 question with minor range despite minor Pass
Noise in Input relatively unaffected
text noise. variations in the
by minor noise.
question text.

44 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

6. Test Cases for Question Paper Generation Module: Test cases for the Question Paper
Generation Module aim to ensure the system's capability to produce high-quality question papers that
align with specified criteria. These cases involve verifying the system's ability to generate questions
covering a range of topics and difficulty levels, ensuring the diversity and balance of question types
(e.g., multiple-choice, short answer), validating the adherence to Bloom's Taxonomy levels in question
selection, and confirming the accuracy of marks allocation.

Table Test Cases for Question Paper Generation Module

Test Stat
Description Steps Expected Result Actual Result
Case ID us

Paper contains 10
1. Provide a CSV file The module will select
Basic unique questions
TC_001 with more than 10 10 random questions Pass
Selection chosen from the CSV
questions. from the file.
file.

The generated papers Paper contains mostly

Selection 1. Generate multiple cannot contain unique question
TC_002 Pass
Uniqueness question papers. identical sets of 10 selections across
questions. multiple generations.

The module will select Paper contains all 10

Limited 1. Provide a CSV file
TC_003 all 10 questions from questions from the CSV Pass
Question Pool with 10 questions.
the file. file.

The module must select Paper contains all

Limited
1. Provide a CSV file all available questions questions from the
Question Pool
TC_004 with less than 10 from the file and inform CSV file with a user Pass
(Less than 10
questions. users about the notification about the
Questions)
limitation. limitation.

The module must return Module identifies an

TC_00 Empty CSV 1. Provide an empty an error message empty file and prevents
Pass
5 File CSV file. indicating the absence paper generation due to
of questions. lack of questions.

The module should Paper is generated

handle the errors using valid questions
1. Provide a CSV
CSV File with gracefully and select from the CSV file, or
file with formatting
TC_006 Formatting valid entries (if any). If an error message is Pass
errors (e.g., missing
Issues the file is unreadable, displayed if the format
or extra columns).
an error message is entirely
should be displayed. incompatible.

45 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

Chapter 10
Results
10.1 Screen shots

1. Dashboard

Fig 10.1.1 : Dashboard

The Question Paper Generator Software appears to be a professional tool designed to help create
customized question papers efficiently. The homepage features a clean, minimalistic layout with a large
open book image in the background, conveying a scholarly and educational tone. The navigation
options at the top of the page, including "Home," "Log In," "Register," and "Template," suggest that
the software provides user accounts, templates, and other features to facilitate the question paper
generation process. This implies the tool is likely intended for use by educators, administrators, or
exam organizers to streamline the creation of customized assessments.

The overall presentation of the website is professional, user-friendly, and focused on providing a
practical solution for the specific needs of its target audience. This software could be a valuable asset
for educational institutions or examination bodies looking to simplify and standardize their question
paper generation workflows.

46 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

2. Registration page

Fig 10.1.1 : Registration page

The registration page serves as the gateway for users to create accounts and gain access to the
question paper generation website. It features intuitive input fields prompting users to provide
essential information such as their full name, email address, and password. The clean, organized
layout and inclusion of a relevant image contribute to a professional and visually appealing user
interface.

The registration process includes validation checks to ensure accurate and secure data entry, such
as the requirement to re-enter the password. This attention to data integrity helps foster trust and
confidence in the platform. Users are encouraged to create strong passwords, prioritizing account
security.

Upon successful registration, users receive a confirmation message and are redirected to the login
page, ready to explore the system's functionalities. The overall design of the registration page is
focused on providing a seamless and user-friendly experience, encouraging engagement with the
platform and its question paper generation capabilities.

47 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

3. Log in page

Fig 10.1.3 : Login page

The login page of the Question Paper Generator platform serves as the secure entry point for registered
users to access the system's features and functionalities. It presents a clean and user-friendly interface,
guiding users through the login process.

The page prominently displays input fields for users to enter their registered email address and
password. This straightforward approach ensures a seamless login experience, allowing authorized
users to quickly and conveniently access the platform.

For users who have not yet registered, the page includes a "Register Now" button, which directs them
to the registration page. This option caters to new users, providing them with a clear pathway to create
an account and gain access to the system.

The login page's design prioritizes both security and user convenience. By requiring valid email and
password credentials, the platform maintains data integrity and ensures that only authorized individuals
can log in. This attention to security helps build user trust and confidence in the system.

Overall, the login page is a crucial component of the Question Paper Generator, offering a professional
and streamlined interface for registered users to access the platform's capabilities. Its intuitive layout
and clear registration option contribute to an inclusive and accessible user experience.

48 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

4. Template page

Fig 10.1.4 : Template page

The Question Paper Generator platform offers a convenient feature that allows users to download a
sample Excel file in CSV format. This file contains pre-populated questions and related information,
serving as a valuable resource for users who need to create custom question papers.

The provided instructions outline a straightforward process for users to create their own CSV file. By
opening a text editor or spreadsheet software, users can simply enter the questions one by one on new
lines, and then save the file with a ".csv" extension. This simplicity ensures an efficient and structured
approach to inputting the necessary content.

Providing this sample file and clear guidelines demonstrates the platform's commitment to streamlining
the question paper creation workflow. This feature caters to the needs of educators, exam
administrators, or content creators, enabling them to quickly and easily customize the questions to suit
their specific requirements.

The inclusion of this downloadable sample file reflects the platform's user-centric design, prioritizing a
seamless and accessible experience for its users. By offering this valuable resource, the Question Paper
Generator aims to empower its audience and facilitate the efficient generation of tailored question
papers.

49 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

5. Generate Question Paper

Fig 10.1.5 : Generate Question Paper page

The question paper generation page provides a user-friendly interface for educators to seamlessly
create customized question papers. The page features a structured layout with various input fields
and dropdown menus to ensure a comprehensive and streamlined experience.

Users can select the college name, branch name, year, and semester from the provided dropdown
options, allowing them to generate question papers tailored to specific academic requirements.
Input fields are available for entering the subject name and faculty details, ensuring accurate
attribution of the generated content.

A calendar dialogue enhances convenience by enabling users to easily select the desired exam date.
This feature helps maintain organization and consistency in the paper creation process.
Additionally, the page includes a designated area for uploading CSV files containing the questions.
This functionality allows educators to efficiently incorporate pre-existing content into the question
paper, further streamlining the overall workflow.

The user-friendly design and comprehensive features of this page empower educators to generate
well-structured question papers that cater to their teaching needs. The intuitive interface and
customization options demonstrate the platform's commitment to providing a seamless and
efficient question paper generation experience.

50 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

5.6 Final Output

Fig 10.1.6 : Result page

The final screenshot showcases a generated question paper that demonstrates the system's ability to
automate the question paper generation process. The paper reflects the selections made by the user in
the previous interface, including the chosen college, branch, year, semester, faculty, subject, and date.

One key advantage of this system is its customizability. Users can easily customize the format of the
question paper, allowing them to tailor the assessment to their specific needs. This could include
adjusting the layout, font, and overall aesthetics of the paper, as well as the distribution of question
types and Bloom's Taxonomy levels.

Furthermore, the system can generate question papers for any number of marks, enabling educators to
create assessments that align with their curriculum and grading schemes. This flexibility ensures that
the generated question papers are not only comprehensive and well-structured, but also adaptable to the
unique requirements of different educational institutions and subject areas.

By automating the question paper generation process and providing customization options, this system
streamlines the assessment creation workflow, saving educators valuable time and effort while
maintaining the integrity and quality of the evaluations. This integration of customization and flexible
mark allocation further enhances the system's capabilities, making it a versatile and efficient tool for
educational institutions.

51 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

Chapter 11
Deployment and Maintenance
11.1 Installation and un-installation
The project requires minimal setup for installation and uninstallation. The steps are as follows:
Installation:
1. Ensure the system meets the hardware and software requirements:
 A modern processor (e.g., Intel Core i3 or equivalent)
 Minimum 4GB RAM for smooth operation
 A compatible operating system (Windows, macOS, or Linux)
 A stable internet connection
2. Install the required software dependencies:
 Python and necessary libraries (NumPy, Pandas, scikit-learn)
 Django web framework
 A web browser for testing and running the interface.
3. Clone or download the project files.
4. Navigate to the project directory and run the setup script using:
python [Link] runserver
5. Open a web browser and access the application through the specified URL.

Uninstallation:
1. Stop the running server.
2. Remove all project-related files from the system.
3. Uninstall any dependencies that were specifically installed for this project.

11.2 User help

To assist users in navigating and using the system efficiently, the following resources are
provided:

1. User Manual – A detailed document explaining how to register, log in, and generate
question papers.
2. FAQ Section – Addresses common queries related to the system's usage.
3. Help Desk Support – Users can contact technical support for assistance with
troubleshooting issues.
4. Tooltips & Error Messages – The system provides real-time guidance through tooltips and
clear error messages to ensure a smooth user experience.

52 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

Chapter 12
Summary and Conclusion
Summary

The project "Bloom's Taxonomy-Based Question Paper Generation Using Machine Learning
Algorithms" introduces an innovative approach to automating the creation of question papers in
educational institutions. By leveraging machine learning algorithms and Bloom’s Taxonomy, this
system categorizes questions based on cognitive complexity, ensuring well-balanced assessments.

Conclusion
In conclusion, our project represents a significant advancement over existing systems, addressing
their limitations while offering a superior solution at an affordable cost. By leveraging Bloom’s
Taxonomy, a renowned framework for educational assessment, we have devised a system capable
of automatically generating high-quality question papers. This system not only creates assessments
but also allocates marks to each question, ensuring accuracy and reliability in evaluation.

Our approach involves categorizing questions based on Bloom’s Taxonomy, allowing for
differentiation across various cognitive levels and educational objectives. The integration of
machine learning further enhances the system's capabilities, enabling it to adapt and improve over
time. With our proposed system, we anticipate not only the correct assignment of marks to
questions but also the generation of multiple question papers tailored to Bloom’s Taxonomy
principles. This innovation promises to streamline the assessment process, promote fairness and
consistency, and ultimately contribute to more effective and efficient educational practices

53 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

CHAPTER 13

REFERENCES

[1] D. V. Paul, “Automatic question paper pattern generation using GA approach,” INFOCOMP
Journal of Computer Science, vol. 19, no. 1, 2020.

[2] M. S. Daware, “Automatic generation of question paper using NLP techniques & taxonomy
rules,” Vidyawarta, 2019.

[3] B. S. A. Shanthi, L. J. R. Harshitha, and K. Manasa, “Automated exam question generator

using genetic algorithm,” 2019.

[4] B. T. G. S. Kumara, A. Brahmana, and I. Paik, “Bloom’s taxonomy and rules based question
analysis approach for measuring the quality of examination papers,” International Journal of
Knowledge Engineering, vol. 5, no. 1, pp. 2–6, 2019.

[5] Y. Timakova and K. A. Bakon, “Bloom’s taxonomy-based examination question paper

generation system,” 2018.

[6] S. V. Zalte, C. C. Jadhav, A. A. Mangire, A. D. Hole, and A. R. Tulshi, “Automatic question

paper generator system,” International Journal of Advanced Research in Computer and
Communication Engineering ISO 3297:2007 Certified, vol. 7, no. 3, 2018.

[7] D. R. Ch and S. K. Saha, “Automatic multiple choice question generation from text: A survey,”
IEEE Transactions on Learning Technologies, 2018.

[8] S. Dhainje, “An automatic question paper generation: using bloom’s taxonomy,” 2018.

[9] A. Killawala, I. Khokhlov, and L. Reznik, “Computational intelligence framework for

automatic quiz question generation,” 978-1-5090-6020-7/18/_c 2018 IEEE.

[10] F. K. Gangar, H. G. Gori, and A. Dalvi, “Automatic question paper generator system,”
International Journal of Computer Applications, vol. 66, no. 10, pp. 42–47, 2017.

[11] Guan, Liancheng. “The design of the automatic generative system of examination papers,”
2017 Sixth International Conference on Future Generation Communication Technologies (FGCT).
IEEE, 2017.

[12] A. S. Khairnar, B. C. Jadhav, R. Birhade, P. Patil, “Automatic question paper generator,”

International Journal For Technological Research In Engineering, vol 4, no. 9, 2017.

[13] A. Leekha, T. Barot, and P. Salunke. “Automatic question paper generator system,”
International Journal of Research in Engineering and Technology.(IJSRET), vol. 6 no. 4, pp. 331–
332, 2017.

54 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

INFORMATION OF PROJECT GROUP

MEMBERS

1. Name : Shruti Jeevan Jadhav

2. Date of Birth : 03-04-2003

3. Gender : Female

4. Permanent Address : Ashtavinayak Society, triveni nagar, Pune 411062

5. E-Mail : [Link]@[Link]

6. Mobile/Contact No :8010497382

7. Paper Published :

55 S.C.O.E., Department of Computer Engineering Year 2024-25

 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

INFORMATION OF PROJECT GROUP

MEMBERS

1. Name : Piyush Premnath Kale

2. Date of Birth : 16-05-2002

3. Gender : Male

4. Permanent Address : Madaj Tal. Umarga Dist. Dharashiv

5. E-Mail : piyushkale454@[Link]

6. Mobile/Contact No. : 9370197097

7. Paper Published :

S.C.O.E., Department of Computer Engineering Year 2024-25

56
 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

INFORMATION OF PROJECT GROUP

MEMBERS

1. Name : Shubham Dnyaneshwar Rajole

2. Date of Birth : 05-09-2004

3. Gender : Male

4. Permanent Address : Karanjgoan, Nashik , Maharashtra, 422210

5. E-Mail : srajole63@[Link]

6. Mobile/Contact No. : 9370308392

7. Paper Published :

S.C.O.E., Department of Computer Engineering Year 2024-25

57
 Blooms Taxonomy Based Question Paper Generation Using Machine Learning Algorithms

INFORMATION OF PROJECT GROUP

MEMBERS

1. Name : Abhishek Premkishor Yadav

2. Date of Birth : 07-02-2002

3. Gender : Male

4. Permanent Address : Pritam prakash nagar, Shirur-Pune 412210

5. E-Mail : yadavprem4605@[Link]

6. Mobile/Contact No. : 9284467538

7. Paper Published :

S.C.O.E., Department of Computer Engineering Year 2024-25

58