You are on page 1of 12

i

“ANOVEL APPROACH FOR SEMANTIC SIMILARITY


MEASUREMENT FOR HIGH QUALITY ANSWER
SELECTION IN QUESTION ANSWERING PLATFORM
USING DEEP LEARNING”

Submitted By
Darshana V. Vekariya
180470702018

Guided By
Prof. Nivid Limbasiya
Assistant Professor,
Computer Engineering Department,
V.V.P. Engineering College,Rajkot

A Thesis Submitted to
Gujarat Technological University in Partial Fulfillment of the Requirements for

the Master of Engineering Degree

June 2020

V.V.P. ENGINEERING COLLEGE


RAJKOT-360005,Gujarat, India

ii
CERTIFICATE

This is to certify that research work embodied in this thesis entitled “A NOVEL
APPROACH FOR SEMANTIC SIMILARITY MEASUREMENT FOR
HIGH QUALITY ANSWER SELECTION IN QUESTION ANSWERING
PLATFORM USING DEEP LEARNING” was carried out by Ms. Darshana
V. Vekariya(Enrollment No:180470702018)at V.V.P. Engineering College,
Rajkot, Gujarat For partial fulfillment of Master of Engineering degree in
Computer Engineering to be awarded by Gujarat Technological University. This
Research work has been carried out under my supervision and is to the
satisfaction of department.

Date:
Place:

Prof. Nivid Limbasiya Dr. Jayesh Deskar


Computer Engineering Department, Principal
V.V.P. Engineering College,Rajkot V.V.P. Engineering College,Rajkot

iii
COMPLIANCE CERTIFICATE

This is to certify that research work embodied in this thesis entitled “A NOVEL
APPROACH FOR SEMANTIC SIMILARITY MEASUREMENT FOR
HIGH QUALITY ANSWER SELECTION IN QUESTION ANSWERING
PLATFORM USING DEEP LEARNING” was carried out by Ms. Darshana
V. Vekariya(Enrollment No:180470702018) at V.V.P. Engineering College,
Rajkot, Gujarat For partial fulfillment of Master of Engineering degree in
Computer Engineering to be awarded by Gujarat Technological University. She
has complied with the comments given by the Dissertation phase-1 as well as
Internal Reviewer to my satisfaction.

Date:
Place:

Darshana V. Vekariya Prof. Nivid Limbasiya


180470702018 Computer Engineering Department,
Computer Engineering Department, V.V.P. Engineering College, Rajkot
V.V.P. Engineering College, Rajkot

iv
PAPER ACCEPTANCE CERTIFICATE

This is to certify that research work embodied in this thesis entitled “A NOVEL
APPROACH FOR SEMANTIC SIMILARITY MEASUREMENT FOR
HIGH QUALITY ANSWER SELECTION IN QUESTION ANSWERING
PLATFORM USING DEEP LEARNING” was carried out by Ms. Darshana
V. Vekariya(Enrollment No:180470702018)at V.V.P. Engineering College,
Rajkot, Gujarat For partial fulfillment of Master of Engineering degree in
Computer Engineering to be awarded by Gujarat Technological University has
published paper entitled A Novel Approach for Semantic Similarity
Measurement for High Quality Answer Selection in Question Answering
System for acceptance by the IEEE 6th International Conference on Advanced
Computing and Communication Systems(ICACCS 2020) on 10 th march 2020.

Date:
Place:

Darshana V. Vekariya Prof. Nivid Limbasiya


180470702018 Computer Engineering Department,
Computer Engineering Department, V.V.P. Engineering College, Rajkot
V.V.P. Engineering College, Rajkot

Dr. Jayesh Deskar


Principal
V.V.P. Engineering College, Rajkot

V
THESIS APPROVAL

This is to certify that research work embodied in this thesis entitled “A


NOVEL APPROACH FOR SEMANTIC SIMILARITY
MEASUREMENT FOR HIGH QUALITY ANSWER SELECTION IN
QUESTION ANSWERING PLATFORM USING DEEP LEARNING” was
carried out by Ms. Darshana V. Vekariya(Enrollment No:180470702018)at
V.V.P. Engineering College, Rajkot, Gujarat For partial fulfillment of
Master of Engineering degree in Computer Engineering to be awarded by
Gujarat Technological University is approved for the degree of Master of
Engineering with specialization in Computer Engineering by Gujarat
Technological University.

Date:
Place:

Examiner’s Sign and Name:

Signature and Name of Examiner Signature and Name of Examiner


Date: Date:

vi
DECLARATION OF ORIGINALITY

I hereby certify that I am the sole author of this thesis and that neither any part of this
thesis nor the whole of the thesis has been submitted for a degree to any other university
or institution.

I certify that, to the best of my knowledge, the current thesis does not infringe upon
anyone’s copyright nor violate any proprietary rights and that any ideas, techniques,
quotations, or any other material from the work of any other people included in our
thesis, published or otherwise, are fully acknowledged in accordance with the standard
referencing practices. Furthermore, to the extent that I have included copyrighted
material that surpasses the boundary of fair dealing within the meaning of the Indian
copyright(Amendment) Act 2012.

I declare that this is true copy of thesis, including any final revisions, as approved by
thesis review committee.

I have checked write up of the present thesis using anti-plagiarism database and it is in
allowable limit. Even though later on this case of any complaint pertaining of plagiarism,
I am sole responsible for the same and I understand that as per UGC norms, University
can even revoke master of engineering degree conferred to the student submitting this
thesis.

Date:

Signature of Student Signature of Guide:


Name of Student : Darshana Vekariya Name of Guide: Prof. Nivid Limbasiya
Enrollment No: 180470702018 Institute Code: 047

vii
ACKNOWLEDMENT

I take this opportunity to express my profound gratitude and deep regards to my Guide
Nivid R. Limbasiya, Assistant professor of Computer Engineering department of V.V.P.
Engineering College, Rajkot, Gujarat for his exemplary guidance, monitoring and
constant encouragement throughout the thesis. Despite his busy schedule, he is always
available to give me advice, support, and guidance during the entire period of my
research work till this date. His insight and creative ideas are always the inspiration for
me during the research.

I wish warmly thanks to Dr. TejasPataliya , Head of Computer Engineering Department


of V.V.P. Engineering College, Rajkot, Gujarat for his cordial support, valuable
guidance and Prof. VirajDaxini, Assistant Professor and PG coordinator for providing
the necessary facilities and guidance during the research and encouragement from time
to time.

I am thankful to all those who directly or indirectly help me in accomplishing my task.


Last, but not least I am grateful thank my institute, V.V.P. Engineering College for
giving me this opportunity to work in the great environment.

Darshana Vekariya
[180470702018]

viii
CONTENTS

Certificate iii
Compliance Certificate iv
Paper Publication Certificate v
Thesis Approval Certificate vi
Statement of Originality vii
Acknowledgements viii
List of Figures xii
List of Tables xii
Abstract xiii
1 Introduction 1
1.1 Overview 1
1.2 Motivation for research 2
1.3 Objective 3
1.4 Applications 4
1.5 Thesis Organization 6

2 Background Study and Literature Review 7


2.1 Overview of deep learning 7
2.2 Overview of CNN 8
2.3 Overview of CNN 9
2.4 Literature Review 11

3. Requirements and Challenges 19


3.1 Research requirement 19
3.2 Challenges 20

4.Implementation Environment 21
4.1 Overview of MATLAB 21
4.2 Local environment setup 23
4.3Deep learning with MATLAB 25

ix
5. Proposed Work 27
5.1 Proposed System 27

6. Result Analysis 38
6.1 Introduction 38
6.2 Result 40
6.3 Comparison with different methods 41

7. Conclusion & Future Work 43


References 44
Appendix A Paper Publication 47
Appendix B Compliance Report 48
Appendix C Review Card 54
Appendix D Plagiarism Report 57

xi
LIST OF FIGURE
Fig No. Name of Figure Pg No.
1.1 Question-Answering System 1
2.1 Deep Learning 7
2.2 Convolution Neural Network 9
2.3 Recurrent Neural Network 10
4.1 MATLAB R2018a 21
4.2 Installation Setup 23
4.3 MATLAB Homepage 23
5.1 Workflow of Proposed Method 27
5.2 CBOW model architecture 31
5.3 Skip-Gram model architecture 31
6.1 Comparison chart 42
A1 Certificate of Published Paper 47
B.1 Modified Proposed Algorithm 50
B.2 Comparison of results 51

LIST OF TABLE
Table No. Name of Table Page
No.
6.1 Result of Proposed system on STSB dataset 42
6.2 Result of Proposed system on SICK dataset 42
6.3 Result of Proposed system on MRPC dataset 42
6.4 Result of Proposed system on Wikipedia dataset 42

6.5 Comparison of different methods with proposed 43


system
A.1 Detail of accepted paper 47

xiii
“A NOVEL APPROACH FOR SEMANTIC
SIMILARITY MEASUREMENT FOR HIGH QUALITY
ANSWER SELECTION IN QUESTION ANSWERING
PLATFORM USING DEEP LEARNING”

Submitted by
DARSHANA V. VEKARIYA
(Enrollment No.180470702018)

Under the guidance of


PROF. NIVID LIMBASIYA
Assistant Professor
Computer Engineering Department
V.V.P. Engineering College, Rajkot

ABSTRACT

Question retrieval and high-quality answer retrieval is the main task in Question-
answering system. It is a real-world application of NLP technologies. The major
challenges of QA is the exact selection of high-quality responses w.r.t. given questions
and by doing that it will also minimize the time of finding a similar question and high-
quality answer. A large amount of data available, but getting the right information
accessible when needed is very important. QA engines attempt to let you ask the
question the way you normally ask. Cosine similarity algorithm is used to compute the
distance between the question and answer pairs. QA can have two domains Open
Domain QA and Closed Domain QA. In Open-domain there is no any particular
domain, users can ask any question, the system will give an answer from the web and
give a respective answer to the user. In a closed domain, QA users can ask a question
related to some particular domain.
Keywords:Question-answering, neural networks, deep learning, memory networks

xiii

You might also like