A

PROJECT SYNOPSIS
On
“An Automatic MCQ & Summary Generation By
Using NLP ”

Submitted By

Name of Student: Roll

No
1) Harshada Jaysing Patil. (20)
2) Mitali Krushnath Dalave. (21)
3) Aarti Sunil Gaikwad. (22)
4) Sneha Mohan Suryawanshi. (23)

Under The Guidance Of

Prof .P. D. Patil

Holywood Academy’s
SANJEEVAN ENGINEERING AND TECHNOLOGY
INSTITUTE, PANHALA
DEPARTMENT OF COMPUTER SCIENCE

&ENGINEERING 2022-2023
 Project Synopsis

Project_Id: 5
Title of Project: An Automatic MCQ & Summary Genration using NLP
Name of Guide: Prof. P. D. Patil
Course Name: Computer Science and Engineering
Name of Institute: Sanjeevan Engineering and Technology Institute, Panhala
Name of Student:

Name of Student: Roll

No
1) Harshada Jaysing Patil (20)
2) Mitali Krushnath Dalave. (21)
3) Aarti Sunil Gaikwad. (22)
4) Sneha Mohan Suryawanshi. (23)

Prof. P. D. Patil Prof. R. S. Nejkar

Project Guide H.O.D.
 INDEX

Sr.no. Content Page no.

1 Introduction 1

2 Literature Survey 2

3 Proposed Work 3

4 Objective 4

5 Modules 5

6 Tool and Technologies 6

7 References 7
 Introduction:

Automatic multiple-choice question generation (MCQG) is a useful still

challenging task in Natural Language Processing (NLP). It is the task of automatic
generation of correct and relevant questions from textual data. Despite its
usefulness, manually creating sizeable, meaningful and relevant questions is a
time-consuming and challenging task for teachers. In the field of education,
questioning is widely acknowledged as an effective instructional strategy to
evaluation or assessment of students or the learner at the end of a lesson.
Questioning as an instructional tool can be traced back to the fourth century,
nowadays most of the exams based on computers in MCQ format.

Natural Language Processing (NLP) is a branch of artificial intelligence that deals

with the interaction between computers and humans using the natural language.
The ultimate objective of NLP is to read, decipher, understand, and make sense of
the written or textual data in natural languages of human in a manner that is
valuable. In this project , we focus on the setting of MCQs for CBTE through NLP
to improve the method of setting MCQs and modification, and for creating a viable
question bank for subsequent use by the academicians for their learners. And
creating the summary of whole paragraph or given data would be easy for quick
understanding and recalling concepts.

This system solves the problem of manual creation of questions and reduces time
consumption and cost.

1|Page
Literature Survey:

[1] The aim of this thesis project was to develop a NLP-based Website for
Teachers to save time and generate questions automatically.
[2] We have studied several research papers on multiple choice question
generation using different approaches. Santhanavijayan. have proposed a
system of “Automatic generation of multiple-choice questions for e-
assessment” . In their proposed system, they have used fireflies-based
preference learning and ontology-based approach to generate MCQs. They
have used a web corpus to make it feasible to create questions.
[3] D. R. CH and S. K. Saha have proposed “Automatic Multiple Choice
Question Generation From Text: A Survey” In this paper, articles from the
database are used to generate questions. NLP-based summarizer is used for
text summarization and frequency count of words, and pattern matching
techniques are used for key selection.
[4] Ayako Hoshino and Hiroshi Nakagawa “A real-time multiple-choice question
generation for language testing: A preliminary study” [2] is based on machine
learning to generate questions automatically. They implement machine
learning algorithms, such as Naive Bayes and KNearest
[5] Deepshree S. Vibhandik et al. have proposed a system, “Automatic / Smart
Question Generation System for Academic Purpose” [4] in which the
Automatic Question Generation system generates specific trigger questions
and multiple-choice questions from student's literature review papers. To
facilitate the generation of specific trigger questions, the system extracts key
concepts from student's papers using the Lingo algorithm.

2|Page
Proposed Work:

The project is based on the generation of MCQG through Natural Language

Processing Techniques. This is used to process the lesson materials/documents fed
by the teacher into a multi-choice questions alongside the answers to each
questions. The NLP processes are applied using TF-IDF and T5 TRANSFORMER.
This system only applies to the text of the document or lesson material in
extracting keywords presented by the teacher. The document is converted into text
file, loaded into the system memory and stored on the system. The text is split into
sentences. The split sentences are tokenized, from which the corpus are built as
TF-IDF and T5 mode. From Figure 1, the raw data is the original text file from the
document (lesson material). Noise removal is any piece of text which is not
relevant to the context of the data such as stop words. The keywords are extracted
by using converts high dimensional features or N different features to the low
dimensional space or 1 feature, it includes tokenization and stemming by
KeyBERT. The cleaned text is the result ready for further experimentation and
such is used to analyse and auto-generate multiple choice questions (MCQs) by
using T5. And by using distractor generation wrong options are generated also with
the summarizer we can create summary of given input Furthermore, we develop a
PDF for easy access and presentation of the MCQs to the teachers for acceptance
or rejection..

3|Page
Objective:

 To replace Manual question paper creation with automatic MCQ generation

system.
 The purpose of An automatic MCQ generation is to reduce human work and
save time of teachers Preparing assignments and assessments is a time-consuming
task for instructors. Automatic Generation (MCQ) was used to create a tool that
automatically generates programming practice exercises thus relieving the instructor
from having to generate them.

4|Page
Modules:
Website Design:
1. Login page.
2. Dropbox layout(for input data
3. Download format (pdf , jpg , word)
4. Edit options.

Data Processing:

First step in module is to load the input data or raw data of any domain for which
questions can be generated.

5|Page
6. Tools And Technologies :
Software Requirment:
 Netbeans:
NetBeans IDE is a free and open source integrated development
environment for application development on windows, Mac, Linux
operating system.
 MySQL:
MySQl is a database management system. To add, access and process data
stored in computer database.
 Xampp Server:
Xmapp is a free and open-source cross-platform web server solution stack
package developed by Apache Friends, consisting mainly of the Apache HTTP
Server, MariaDB database, and interpreters for scripts written in the php and
Perl programming languages.
 Android Studio:
Android Studio is the official Integrated Development Environment for
Android app development, based on IntelliJ IDEA. A unified environment
where you can develop for all android devices.

Languages Used
HTML, CSS, Javascript, PHP, JAVA

Hardware requirement:
Laptop/Computer
RAM : 8GB

6|Page
7|Page
7. References:
[1] Published in: Bello Ahmed Dangiwa and Smitha S Kumar, “An OCR system
for business cards[IEEE]”, INSPEC 18471887, 2019,link:
https://ieeexplore.ieee.org/document/8642727.
[2] Published in: H. Saiga and Y. Nakamura, “An OCR system for businesscards
[IEEE]”, INSPEC4951050, 2002, link:
https://ieeexplore.ieee.org/document/395616.
[3] Published in: Xuewen Zhao and Qiang Gao, “Business Card Recognition and
E- mail Delivery Based on NAO Robot[IEEE]”, INSPEC 17897852,2018, Link:
https://ieeexplore.ieee.org/document/8408107
[4] Published in : Alok Patel. 2012 Lets compare applets and oranges: NFC
versus QR code. Available at:http://www.atuch.com/blogs/news/5644452-lets-
compare- applets-oranges-nfc-versus-qr-code-comparison-table-what-shall-you-
pick[Accessed 25 march 2013].
[5] Published in: Prof. Kaushal Patel, Prof. Amit Choksi.2016 Android Based
Business Card Scanner:An OCR Based Approach. Link: http: //ijcsit.com /docs/
Volume% 207/ vol7issue1/ijcsit2016070176.pdf

8|Page