IRJET - Automated System for Frequently Occurred Exam Questions
0 likes•3 views
This document summarizes an automated system for generating exam question banks. It discusses the limitations of the traditional manual method for creating question banks, such as being time-consuming and prone to errors. The proposed system aims to automatically generate question banks from existing question papers in a database. It uses information retrieval and natural language processing techniques like stemming, stopword removal, and text ranking to analyze question papers according to the syllabus. This allows question banks to be generated quickly and reliably. The system is intended to save lecturers time while improving the quality and flexibility of question banks. It also discusses the workflow and advantages of the proposed automated system over the existing manual process.
![International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 07 Issue: 02 | Feb 2020 www.irjet.net p-ISSN: 2395-0072
© 2020, IRJET | Impact Factor value: 7.34 | ISO 9001:2008 Certified Journal | Page 3014
Automated System for Frequently Occurred Exam Questions
Manjusha sanke1, Nihar Pednecar2, Viraj Padwalkar2, Raghavendra Poojary2, Vikas
Vishwakarma2
1Assistant Professor, Department of Information Technology, Shree Rayeshwar Institute of Engineering and
Information Technology Goa, India
2Department of Information Technology, Shree Rayeshwar Institute of Engineering and Information Technology
Goa, India
---------------------------------------------------------------------***----------------------------------------------------------------------
Abstract - Education has become an integral part of our
society today. This is the reason why examinations play an
important role in testing the performance of the student.
Generating an effective question bank is a task of great
importance for any educational institute. The traditional
method where lecturers and students manually prepare
question bank, is very challenging, timeconsumingandmostly
inaccurate. In this paper, we presentthesolutioninthe form of
Automated System for Frequently Occurred Exam Questions.
This web based application will enable lecturers to
automatically generate question bank from existing question
papers in the database in a very short time and also help
students.
Key Words: component, formatting, style, styling, insert
(key words)
1. INTRODUCTION
There is a change from manual to automated systems for
various fields of education system by student as well as
teachers. However the main problem is a low quality of
question banks generated because of some human factors
such as instability and relatively narrow scope of topics.
Teachers need to invest a lot oftimeandenergyincomposing
question banks. Also, the concern remains is how thecurrent
technologies would also help the lecturers automatically
generate the different sets of questions from time to time
without being concern about repetition. Using this system,
the same action can be accomplished in minutes.
According to the need, a system named as Automated
System for FrequentlyOccurredExamQuestionsisproposed,
so as to make the system more efficient, reliable, improve its
quality, and also to reduce the time taken by lecturers in
setting the question bank manually. This web based
application will enable lecturers to automatically generate
question bank from existing question papers in thedatabase.
The system performs all tasks related to question bank,
starting from preparing question bank to printing it. This
system is very useful for students as well as lecturers. The
system will help the individuals to generate question bank in
a very short time period thus saving a lot of their precious
time.
2. LITERATURE SURVEY
2.1. Information Retrieval System
Information Retrieval is a field of computer science that
deals with the representation, storage and access of
information. Information Retrieval deals with organization
and retrieval of information from database collections [6].IR
is the process by which a collection of data is searched,
represented and stored for the purpose of knowledge
discovery as a response to a user's request.
IR is a component of an information system. An
information system must make sure that everybody it is
meant to serve has the information needed to accomplish
tasks, solve problems and make decision, no matter where
the information is available. An information retrieval system
mustactively find out whatusersneed,acquiredataresulting
in a collection and match data with needs. Figuring out the
information that the user needs to solve a problem is
important for successful retrieval.
2.2. Information Retrieval Model
IR models[8] specify the details of the document
representation, the query representation and the retrieval
functionality. Set-theoretic, algorithm, probabilistic, feature
based models are the different models in IR.
2.2.1. Set-Theoretic model
Set-theoretic models represent documents as sets of
words or phrase. Similarities are usually computed from set-
theoretic operations on those sets. Commons models are:
Standard Boolean model
Extended Boolean model
Fuzzy retrieval
2.2.2. Probabilistic model
Probabilistic models[3] make the process of document
retrieval as a probabilisticinference.Similaritiesareobtained
as probabilities that a document is related for a given query.](https://image.slidesharecdn.com/irjet-v7i2646-201128045752/85/IRJET-Automated-System-for-Frequently-Occurred-Exam-Questions-1-320.jpg)
![International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 07 Issue: 02 | Feb 2020 www.irjet.net p-ISSN: 2395-0072
© 2020, IRJET | Impact Factor value: 7.34 | ISO 9001:2008 Certified Journal | Page 3015
Probabilistic theoremsliketheBayes’theoremareoftenused
in these models.
Probabilistic relevance model on which it is based
the okapi (BM25) relevance function.
Uncertain inference
Language models
Divergence-from-randomness model
Latent Dirichlet allocation
Binary Independence model
2.2.3. Algebraic model
Algebraic models represent document and queries
usually as vectors or tuples. The similarity between the
query vector and document vector is represented as a scalar
value.
Vector space model
Generalized vector space model
(Enhanced) Topic-based Vector Space Model
Extended Boolean model
Latent semantic indexing
Vector Space Model
The fundamental IR model used in our system is vector
space model. A vector space model is also an algebraic
model, involving two steps:
In first step we represent the text documentinto
vector of words.
In second step we change to number format so
that we can apply any text mining techniques
such as information retrieval.
In a Vector Space Model
Document and queries are M-dimensional term
vectors.
Non binary weights to index terms.
A query is similar to document if their vectors
are similar.
Retrieved documents are sorted by decreasing
order.
2.3. Implementation Techniques
2.3.1. Data Pre-processing Algorithm
2.3.1.1. Stemming
Removing suffixes by automatic manner is an operation
which is especially useful inthearea ofinformationretrieval.
In a typical IR environment, one has a huge collection of
documents, each described by the words in the document
title and by words in the document abstract. [9]
Ignoring the problem ofpreciselywherethewordsoriginate,
we can say that a document is represented by a vector of
words or terms. Terms with a common stem will usually
have same meanings[1], for example:
CONNECT
CONNECTED
CONNECTING
CONNECTION
CONNECTIONS
The performance of an IR system will be enhanced if term
groups such as this are conflated into a singleterm.Thismay
be done by removing the suffixes
-ED, -ING, -ION, -IONS to leave the single term CONNECT. In
addition, the suffix stripping process will reduce the total
number of terms in the IR system and hence the size and
complexity of the data in the system, which is always
advantageous.
2.3.1.2. Stop word Removal
Stop-words[5] are the words which do not add much
meaning to a sentence. They can safely be ignored without
sacrificing the meaning of the sentence.
You should remove these tokens only if they don’t add any
new information to the sentence. Classification problems
normally don’t need stop words because it is possible totalk
about the general idea of a text even if you remove stop
words from it.
The algorithm is implemented as below given steps:
Step 1: The document from which stop words have to be
removed that target document has to be tokenized
and than individual words are stored in array.
Step 2: A single stop word is read from stop word list one by
one.
Step 3: The stop word is compared to target document text
in the form of array using sequential search
technique.
Step 4: If it matches, the word in array is removed and the
comparison is continue till length of array.
Step 5: After removal of stop words completely,anotherstop
words is read from stop word list andagainalgorithm
follows Step 2. The algorithm runs continuously until
all the stop words are compared and removed.
Step 6: Resultant text after removing stop words is
displayed, also required statistics like stop word
removed, number of stop words removed fromtarget
text, total count of words in target text, count of](https://image.slidesharecdn.com/irjet-v7i2646-201128045752/85/IRJET-Automated-System-for-Frequently-Occurred-Exam-Questions-2-320.jpg)
![International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 07 Issue: 02 | Feb 2020 www.irjet.net p-ISSN: 2395-0072
© 2020, IRJET | Impact Factor value: 7.34 | ISO 9001:2008 Certified Journal | Page 3016
words in resultant text, individual stop word count
found in target document text is displayed.
2.3.1.3. Text Rank
Text Rank[2] is an extractive and unsupervised text
summarization technique. The algorithm works as follows:
The first step is to concentrate all the text
containing in the articles.
Then split the text into individual sentence.
Next, we will find vector representation (word
embeddings) for each and every sentence.
Similaritiesbetweensentencevectorsarecalculated
and stored in a matrix.
The similarities matrix is then converted into a
graph, with sentence as vertices and similarity
scores as edges, for sentence rank calculation.
At last a certain number of top-ranked sentences
form the final summary.
3. Application
3.1. Existing System
The existing system is paper based system.
The production of question bank is done manually.
It needs human working to create question bank
.For this teachers select questions by checking
syllabus and previous question paper.
Drawbacks of Existing System:-
Manual process consumes more time.
More time and effort require managing and
retrieving from manual question papers.
It is difficult to handle with repetition.
3.2. Proposed System
The proposed system will overcome limitations of
existing.
Automatic generation of question bank.
Reduce time and efforts.
Provides moresecurityandflexibilityforgeneration
of question bank.
Advantages of proposed system:-
The entry of the questions is all done by facilities
with syllabus and subject label.
The user can take print of this questionbank orthey
can save it for future use.
This system will help us to keep ourdata integrated.
The problem of data loss will be eliminated.
3.3 Working of the System
User uploads a particular subject syllabus file.
After the syllabus is uploaded it will be convertedin
the form of query. This query will be used to select
the question papers of the particular subject from
the corpus.
Set of question papers will be stored in a data set in
the system.
The question papers of a particular subject will be
fetched by the query for analysing the question
papers.
The system will perform the analysis on the
question papers according to the syllabus.
Fig.1. Working of the system
4. CONCLUSION
In this paper, a new automated system for frequently
occurred exam questions has been proposed. When we
compare the proposed system with the existing system, we
found that the proposed system is more reliable, accurate
and feasible in comparison with existing system. The
proposed system saves lot of time with very less effort. In
future we will enhance the scope by creating the list of
important topics for semester exams or university exam.
REFERENCES
1. Ms. Anjali Ganesh Jivani, “A ComparativeStudyofStudy
Stemming Algorithm”. Anjali Ganesh Jivani et al, Int. J.
Comp. Tech. Appl., Vol 2 (6), 1930-1938
2. Rada Mihalcea and Paul Tarau, “TextRank: Bringing
Order into Texts”
3. Akram Roshdil and Akram Roohparvar, “Review:
Information Retrieval Techniques and Application”.
International Journal of Computer Networks and
Communications Security VOL.3,no.9,September2015,
373-377](https://image.slidesharecdn.com/irjet-v7i2646-201128045752/85/IRJET-Automated-System-for-Frequently-Occurred-Exam-Questions-3-320.jpg)
IRJET - Automated System for Frequently Occurred Exam Questions
- 1. International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056 Volume: 07 Issue: 02 | Feb 2020 www.irjet.net p-ISSN: 2395-0072 © 2020, IRJET | Impact Factor value: 7.34 | ISO 9001:2008 Certified Journal | Page 3014 Automated System for Frequently Occurred Exam Questions Manjusha sanke1, Nihar Pednecar2, Viraj Padwalkar2, Raghavendra Poojary2, Vikas Vishwakarma2 1Assistant Professor, Department of Information Technology, Shree Rayeshwar Institute of Engineering and Information Technology Goa, India 2Department of Information Technology, Shree Rayeshwar Institute of Engineering and Information Technology Goa, India ---------------------------------------------------------------------***---------------------------------------------------------------------- Abstract - Education has become an integral part of our society today. This is the reason why examinations play an important role in testing the performance of the student. Generating an effective question bank is a task of great importance for any educational institute. The traditional method where lecturers and students manually prepare question bank, is very challenging, timeconsumingandmostly inaccurate. In this paper, we presentthesolutioninthe form of Automated System for Frequently Occurred Exam Questions. This web based application will enable lecturers to automatically generate question bank from existing question papers in the database in a very short time and also help students. Key Words: component, formatting, style, styling, insert (key words) 1. INTRODUCTION There is a change from manual to automated systems for various fields of education system by student as well as teachers. However the main problem is a low quality of question banks generated because of some human factors such as instability and relatively narrow scope of topics. Teachers need to invest a lot oftimeandenergyincomposing question banks. Also, the concern remains is how thecurrent technologies would also help the lecturers automatically generate the different sets of questions from time to time without being concern about repetition. Using this system, the same action can be accomplished in minutes. According to the need, a system named as Automated System for FrequentlyOccurredExamQuestionsisproposed, so as to make the system more efficient, reliable, improve its quality, and also to reduce the time taken by lecturers in setting the question bank manually. This web based application will enable lecturers to automatically generate question bank from existing question papers in thedatabase. The system performs all tasks related to question bank, starting from preparing question bank to printing it. This system is very useful for students as well as lecturers. The system will help the individuals to generate question bank in a very short time period thus saving a lot of their precious time. 2. LITERATURE SURVEY 2.1. Information Retrieval System Information Retrieval is a field of computer science that deals with the representation, storage and access of information. Information Retrieval deals with organization and retrieval of information from database collections [6].IR is the process by which a collection of data is searched, represented and stored for the purpose of knowledge discovery as a response to a user's request. IR is a component of an information system. An information system must make sure that everybody it is meant to serve has the information needed to accomplish tasks, solve problems and make decision, no matter where the information is available. An information retrieval system mustactively find out whatusersneed,acquiredataresulting in a collection and match data with needs. Figuring out the information that the user needs to solve a problem is important for successful retrieval. 2.2. Information Retrieval Model IR models[8] specify the details of the document representation, the query representation and the retrieval functionality. Set-theoretic, algorithm, probabilistic, feature based models are the different models in IR. 2.2.1. Set-Theoretic model Set-theoretic models represent documents as sets of words or phrase. Similarities are usually computed from set- theoretic operations on those sets. Commons models are: Standard Boolean model Extended Boolean model Fuzzy retrieval 2.2.2. Probabilistic model Probabilistic models[3] make the process of document retrieval as a probabilisticinference.Similaritiesareobtained as probabilities that a document is related for a given query.
- 2. International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056 Volume: 07 Issue: 02 | Feb 2020 www.irjet.net p-ISSN: 2395-0072 © 2020, IRJET | Impact Factor value: 7.34 | ISO 9001:2008 Certified Journal | Page 3015 Probabilistic theoremsliketheBayes’theoremareoftenused in these models. Probabilistic relevance model on which it is based the okapi (BM25) relevance function. Uncertain inference Language models Divergence-from-randomness model Latent Dirichlet allocation Binary Independence model 2.2.3. Algebraic model Algebraic models represent document and queries usually as vectors or tuples. The similarity between the query vector and document vector is represented as a scalar value. Vector space model Generalized vector space model (Enhanced) Topic-based Vector Space Model Extended Boolean model Latent semantic indexing Vector Space Model The fundamental IR model used in our system is vector space model. A vector space model is also an algebraic model, involving two steps: In first step we represent the text documentinto vector of words. In second step we change to number format so that we can apply any text mining techniques such as information retrieval. In a Vector Space Model Document and queries are M-dimensional term vectors. Non binary weights to index terms. A query is similar to document if their vectors are similar. Retrieved documents are sorted by decreasing order. 2.3. Implementation Techniques 2.3.1. Data Pre-processing Algorithm 2.3.1.1. Stemming Removing suffixes by automatic manner is an operation which is especially useful inthearea ofinformationretrieval. In a typical IR environment, one has a huge collection of documents, each described by the words in the document title and by words in the document abstract. [9] Ignoring the problem ofpreciselywherethewordsoriginate, we can say that a document is represented by a vector of words or terms. Terms with a common stem will usually have same meanings[1], for example: CONNECT CONNECTED CONNECTING CONNECTION CONNECTIONS The performance of an IR system will be enhanced if term groups such as this are conflated into a singleterm.Thismay be done by removing the suffixes -ED, -ING, -ION, -IONS to leave the single term CONNECT. In addition, the suffix stripping process will reduce the total number of terms in the IR system and hence the size and complexity of the data in the system, which is always advantageous. 2.3.1.2. Stop word Removal Stop-words[5] are the words which do not add much meaning to a sentence. They can safely be ignored without sacrificing the meaning of the sentence. You should remove these tokens only if they don’t add any new information to the sentence. Classification problems normally don’t need stop words because it is possible totalk about the general idea of a text even if you remove stop words from it. The algorithm is implemented as below given steps: Step 1: The document from which stop words have to be removed that target document has to be tokenized and than individual words are stored in array. Step 2: A single stop word is read from stop word list one by one. Step 3: The stop word is compared to target document text in the form of array using sequential search technique. Step 4: If it matches, the word in array is removed and the comparison is continue till length of array. Step 5: After removal of stop words completely,anotherstop words is read from stop word list andagainalgorithm follows Step 2. The algorithm runs continuously until all the stop words are compared and removed. Step 6: Resultant text after removing stop words is displayed, also required statistics like stop word removed, number of stop words removed fromtarget text, total count of words in target text, count of
- 3. International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056 Volume: 07 Issue: 02 | Feb 2020 www.irjet.net p-ISSN: 2395-0072 © 2020, IRJET | Impact Factor value: 7.34 | ISO 9001:2008 Certified Journal | Page 3016 words in resultant text, individual stop word count found in target document text is displayed. 2.3.1.3. Text Rank Text Rank[2] is an extractive and unsupervised text summarization technique. The algorithm works as follows: The first step is to concentrate all the text containing in the articles. Then split the text into individual sentence. Next, we will find vector representation (word embeddings) for each and every sentence. Similaritiesbetweensentencevectorsarecalculated and stored in a matrix. The similarities matrix is then converted into a graph, with sentence as vertices and similarity scores as edges, for sentence rank calculation. At last a certain number of top-ranked sentences form the final summary. 3. Application 3.1. Existing System The existing system is paper based system. The production of question bank is done manually. It needs human working to create question bank .For this teachers select questions by checking syllabus and previous question paper. Drawbacks of Existing System:- Manual process consumes more time. More time and effort require managing and retrieving from manual question papers. It is difficult to handle with repetition. 3.2. Proposed System The proposed system will overcome limitations of existing. Automatic generation of question bank. Reduce time and efforts. Provides moresecurityandflexibilityforgeneration of question bank. Advantages of proposed system:- The entry of the questions is all done by facilities with syllabus and subject label. The user can take print of this questionbank orthey can save it for future use. This system will help us to keep ourdata integrated. The problem of data loss will be eliminated. 3.3 Working of the System User uploads a particular subject syllabus file. After the syllabus is uploaded it will be convertedin the form of query. This query will be used to select the question papers of the particular subject from the corpus. Set of question papers will be stored in a data set in the system. The question papers of a particular subject will be fetched by the query for analysing the question papers. The system will perform the analysis on the question papers according to the syllabus. Fig.1. Working of the system 4. CONCLUSION In this paper, a new automated system for frequently occurred exam questions has been proposed. When we compare the proposed system with the existing system, we found that the proposed system is more reliable, accurate and feasible in comparison with existing system. The proposed system saves lot of time with very less effort. In future we will enhance the scope by creating the list of important topics for semester exams or university exam. REFERENCES 1. Ms. Anjali Ganesh Jivani, “A ComparativeStudyofStudy Stemming Algorithm”. Anjali Ganesh Jivani et al, Int. J. Comp. Tech. Appl., Vol 2 (6), 1930-1938 2. Rada Mihalcea and Paul Tarau, “TextRank: Bringing Order into Texts” 3. Akram Roshdil and Akram Roohparvar, “Review: Information Retrieval Techniques and Application”. International Journal of Computer Networks and Communications Security VOL.3,no.9,September2015, 373-377
- 4. International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056 Volume: 07 Issue: 02 | Feb 2020 www.irjet.net p-ISSN: 2395-0072 © 2020, IRJET | Impact Factor value: 7.34 | ISO 9001:2008 Certified Journal | Page 3017 4. Jovdip Datta, “Ranking in Information Retrieval”. April 16, 2010 5. Balasubramai Ramasamy, “Survey on Pre-processing Techniques for Text Mining ”, June 2016 6. Ricardo A. Baeza-Yates and Berthier Ribeiro-Neto. Modern Information Retrieval. Addison-Wesley Longman Publishing Co., Inc., Boston,USA, 1999. 7. Vaibhav Kant Singh, Vinay Kumar Singh "Vector Space Model: An Information Retrieval System” 8. Hai Dong, Farookh KhadeerHussain,ElizabethChang“A Survey in Traditional Information Retrieval Models”, 2008 Second IEEE International Conference on Digital Ecosystems and Technologies. 9. Martin F. Porter. An algorithm for suffix stripping. Pages 313–316, 1997. 10.B. A. Ribeiro-Neto and R. Muntz, "A brief network model for IR," in 19th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, Zurich, 1996, pp. 235-260. 11.Vikram Singh, Balwinder Saini, “An Effective Pre- Processing Algorithm for Information Retrieval Systems”, January2015DOI: 10.5121/ijdms.2014.6602