DEVELOPING A FRAMEWORK FOR ONLINE PRACTICE EXAMINATION AND AUTOMATED SCORE GENERATION

International Journal of Computer Science & Information Technology (IJCSIT) Vol 10, No 6, December 2018
DOI: 10.5121/ijcsit.2018.10602 13
DEVELOPING A FRAMEWORK FOR ONLINE
PRACTICE EXAMINATION AND AUTOMATED SCORE
GENERATION
S. M. Saniul Islam Sani, Rezaul Karim and Mohammad Shamsul Arefin
Department of Computer Science & Engineering, Chittagong University of Engineering
& Technology, Chittagong-4349, Bangladesh
ABSTRACT
Examination is the process by which the ability and the quality of the examinees can be measured. It is
necessary to ensure the quality of the examinees. Online examination system is the process by which the
participants can appear at the examination irrespective of their locations by connecting to examination site
via Internet using desktop computers, laptops or smart phones. Automated score generation is the process
by which the answer scripts of the examinations are evaluated automatically to generate scores. Although,
there are many existing online examination systems, the main drawback of these systems is that they cannot
compute automated score accurately, especially from the text-based answers. Moreover, most of them are
unilingual in nature. As a result, examinees can appear at the examination in a particular language.
Considering this fact, in this paper, we present a framework that can take Multiple Choice Questions
(MCQ) examinations and written examinations in two different languages English and Bangla. We develop
a database where the questions and answers are stored. The questions from the database are displayed in
the web page with answering options for the MCQ questions and text boxes for the written questions. For
generating the scores of the written questions, we performed several types of analysis of the answers of the
written questions. However, for generating the scores of the MCQ questions, we simply compared between
the database answers and the user’s answers. We conducted several experiments to check the accuracy of
score generation by our system and found that our system can generate 100% accurate scores for MCQ
questions and more than 90% accurate scores from text based questions.
KEYWORDS
Multiple choice questions, automated scoring, answer analysis, experimental analysis.
1. INTRODUCTION
Examination system is helpful to separate teaching and testing. It promotes teachers to make the
lesson planning and teach in the class carefully. It is also an important mean of evaluating the
teaching effect. Online practice examination systems can play an important role in the education
system. Introduction of online examination systems can save time and money as well as it can
justify ones capability. Though online examination it is not yet very popular in our country, it
seems to be very useful for the students to take preparations for different examinations.
Considering this fact, we have developed an online examination system, where students can
appear different types of MCQ and written examinations and it is possible to generate automatic
score when examination is taken. If one has computer with Internet connection, he / she can give
the examination in real time and get the results of the examination instantly. He / she can even
appear at the examination while moving using a laptop or a smart phone.

14
The main objectives of our work are illustrated as follows:
• To develop a framework for online practice examination with automated score
generation.
• To develop a user friendly interface for computers and smart phones so that the students
can appear at the examination from easily.
The main contributions of this paper can be summarized as follows:
• We develop a system for online practice examination for the university entrance
examinations.
• We develop a question database from which questions are generated dynamically.
• We evaluated answers to generate scores for both MCQ and written examinations.
The remainder of the paper is organized as follows. Section 2 provides a brief review of related
works. Section 3 describes details about the architecture and methodology of our system. Section
4 gives the experimental results. Finally, we conclude the paper in Section 5. This document
describes, and is written to conform to, author guidelines for the journals of AIRCC series. It is
prepared in Microsoft Word as a .doc document. Although other means of preparation are
acceptable, final, camera-ready versions must conform to this layout. Microsoft Word
terminology is used where appropriate in this document. Although formatting instructions may
often appear daunting, the simplest approach is to use this template and insert headings and text
into it as appropriate.
2. RELATED WORK
Web-based examination system (Fang et al. , 2012) provides the functionality of question
management, paper generation and online tests.
two models are designed to develop a system by which online examination can be held. Here how
a teacher set a question paper and then test would hold by online. But no automated scoring is
possible. In 1966, Ellis Page showed that an automated “rater” is indistinguishable from human
raters [2].
Latent semantic analysis (LSA) is a technique in natural language processing, in particular in
vectorial semantics, of analyzing relationships between a set of documents and the terms they
contain by producing a set of concepts related to the documents and terms. LSA assumes that
words that are close in meaning will occur in similar pieces of text. A matrix containing word
counts per paragraph (rows represent unique words and columns represent each paragraph) is
constructed from a large piece of text and a mathematical technique called singular value
decomposition (SVD) is used to reduce the number of rows while preserving the similarity
structure among columns [3], [9]. Page developed PEG the first attempt at scoring essays by
computer. Page uses the terms trins and proxes for grading an essay; trins refer to the intrinsic
variables such as fluency, diction, grammar, punctuation, etc., proxes denote the approximation
(correlation) of the intrinsic variables.
IntelliMetric [4] uses a blend of Artificial Intelligence (AI), Natural Language Processing (NLP),
and statistical technologies. Search is a system specifically developed for use with IntelliMetric to
understand natural language to support essay scoring. IntelliMetric needs to be “trained” with a
set of essays that have been scored beforehand including “known scores” determined by human
expert raters.

15
The various AES programs differ in what specific surface features they measure, how many
essays are required in the training set, and most significantly in the mathematical modeling
technique. Early attempts used linear regression. Modern systems may use linear regression or
other machine learning techniques often in combination with other statistical techniques such as
Linear Semantic Analysis and Bayesian inference [5].
There are many limitations in the above related works. Most of the automated score is done on
the basis of English literature and essay score. There are some few works on the automated score
for the online examination for mathematics and physics. It is challenging to score the answer
script automated for more than one subjects.
3. SYSTEM DESIGN AND IMPLEMENTATION
The online practice examination system is a client server model. In the system, web browser is
used as client, JSP Engine as the business logic tier to achieve its function, and database system
as the data layer. Client is web browser, which is implemented the system's display logic. The
function sent request to Web Server through the Web browser by users.
Figure 1: Flow diagram of the system
While Web Server return the requested HTML pages or HTML pages dynamically generate by
JSP page to the client, which are shown in the Web browser. Business logic tier is achieved

16
mainly by JSP and Java Bean running in the JSP Engine. It responds to client requests and
achieved the business logic with the Web Server. Tomcat, open source software is used as the JSP
Engine and Web Server. Data tier is realized with database system, is used to store the business
data such as questions and papers and control data such as user data. MS ACCESS is used to
achieve the data tier. The flow graph of the proposed method is given in Figure 1.
The user enters to the system from the URL: bdadmissionexam.com and selects the type of the
examination. When the user selects the admission type, he will route into another web page. If
he/she selects the University or Medical option he/she can answer only MCQ type questions. If
he/she selects the Engineering option he/she will get both MCQ and written questions. Figure 2
shows the procedure for the selection of examination type.
Figure 2: Procedure for selection of examination type
In both written and MCQ type questions, the users can select selective type of subjects to participate in the
practice exam. In case of university type, the subjects are Physics, Chemistry, Mathematics, English and
Biology. In case of Engineering type, the subjects for both written and MCQ questions are Physics,
Chemistry, Mathematics and English. In case of Medical type the subjects are Biology, Physics, Chemistry
and English. Figure 3 shows the code segment for subject selection.
Figure 3: Procedure for selection of subject
Figure 3: Procedure for selection of subject
When the user selects a subject for a particular type then a question paper is displayed on the web
page. The question generation is changeable i.e. question once appeared first time may or may not
be in the second or other time. The database for the questions of different subjects is created with
the appropriate answer of the questions. The database is created by MySQL and PHP is used to
Algorithm select_Admission_Type Method:
POST
action='route.php'
1. switch select from name="sub"
2. case "eng":Engineering
3. case "uni":University
4. case "med":Medical
</select>
<input type='submit' name='submit'>
Algorithm: MCQ_Exam
Method=POST
//Select Any Subject:
1. switch select from name="sub"
2. case "biology":Biology
3. case "chem":Chemistry
4. case "english":English

17
generate the question for the students. An answering block will be there for the written questions.
For the MCQ questions the four options will displayed with radio button. Only one option can
select which should be the correct answer. Algorithm in Figure 4 performs the task.
Figure 4: Procedure for MCQ question pattern
When the question paper is displaying then a timer is set for each subject and when the time is
finished the question paper will submit automatically. If the students want to submit the answer
before time then he/she will select the answer button. Programming language Ajax is used to
count time during the examination. The algorithm is shown in Figure 5.
Figure 5: Procedure for timer control
When the user finishes the examination he/she will select the answer button or it will
automatically submit after the time finishes. Then it is ready to score. The answer collect from
the user is compare with the answer from the database. We use similar-text algorithm for
comparing written examination. This calculates the similarity between two strings as described in
Programming Classics: Implementing the World's Best Algorithms by Oliver. Note that this
implementation does not use a stack as in Oliver's pseudo code, but recursive calls which may or
may not speed up the whole process.
As the users attend the examination they will want to see the result of the examination very
quickly. This procedure is shown in Figure 6.
Algorithm: Displayig_question
1. while found until row =
mysqli_fetch_array(result)
2. show iteration.".".row['question']
3. put id=row['id']
4. put a1=$row['a1']
5. put a2=row['a2']
6.put a3=row['a3']
Algorithm: Start_Time
function Decrement() {
1. initiate currentMinutes = Math.floor(secs / 60);
2. initiate currentSeconds = secs % 60;
3. if currentSeconds less than or equal 9
4. do currentSeconds = "0" + currentSeconds;

18
Figure 6: Procedure answer evaluation
But before displaying the result the score will be generated and counted. After the comparing the
user answer with the appropriate answer in the database the score is generated. The following
algorithm is required to generate score of the examination.
Figure 7: Procedure for score generation of MCQ type questions
Generation of score for MCQ type is entirely different from the written type. The answer will
compare with the answer database and the correct answers will count and add with the total score.
The following algorithm is used to count score of written questions.
Figure 8: Procedure for score generation of written type questions
Algorithm: Generate_Score
1. For each POST as keys=>val
2. keys;
3. val;
4. Givemark ($keys,$val,$sub);
5. endfor
Algorithm: Count_Score
function givemark(keys,val,sub){
1. input: keys, val,sub
2. result = mysqli_query(con,"SELECT * FROM
sub where id='keys' AND answer LIKE '%val%'");
3. rowcount=mysqli_num_rows(result);
4. rowcount;
5. if $rowcount greater than 0
6. Increment GLOBALS['mark']
7. endif}
Algorithm: Evaluate_Answer
1. while found until row =
mysqli_fetch_array(result){
2. initiate ans=row['answer'];
3. val;
4. similar_text(ans, val, p);
5. put p=(p/10);
}

19
4. EXPERIMENTS
The experimental result based on some examinations given by some of the CUET students. They
attended the examinations which were graded by our system. The same answers were scored
manually. Then the deviation of the automated score and the human score were computed and the
accuracy of the developed system were measured.
In order to take the examination, five students from different halls of a public university were
chosen. We considered five different question sets for each subject and assigned a separate
question set to each of the five students. Each answer script of written type examination was
scored by our developed system and by a human expert. Then, we compared human score and
automated score as shown in Figure 9 to Figure 12 for different subjects.
Figure 9: Human & system score for physics
Figure 10: Human & system score for chemistry

20
Figure 11: Human & system score for mathematics
Figure 12: Human & system score for English
5. CONCLUSION
Online practice examination and automated score will help the students to prepare themselves for
the admission examination in different universities. By this system the students can prepare
themselves while staying in their houses as well as while they are moving with their laptops and
smart phones. As the system is capable of generating automated score so the students will find
themselves beneficial. We tested our system against human generated score and found that our
system is an efficient one.
REFERENCES
[1]. Qiao-fang, Z., and L. Yong-fei, 2012. Research and development of online examination system. 2nd
International Conference on Computer and Information Application (ICCIA 2012), pp. 1-3.
[2]. S. Valenti, F. Neri, and A. Cucchiarelli, “An overview of current research on automated essay
grading,” Journal of Information Technology Education, vol. 2, pp.319-330,2003.
[3]. T. Miller, “Essay assessment with latent semantic analysis,” Department of Computer Science,
University of Toronto, Toronto, ON M5S 3G4, Canada, 2002.

21
[4]. L. M. Rudner, V. Garcia, and C. Welch, “An evaluation of the IntelliMetric essay scoring system,”
The Journal of Technology, Learning, and Assessment, vol. 4, no. 4, pp. 1-22, March 2006.
[5]. Rudner, Lawrence (ca. 2002). “Computer Grading using Bayesian Networks-Overview”. Retrieved
2012-03-07
[6]. FAN Ming-hu, SUN Bin “Design and implementation of general test questions library management
system” Computer engineering and Design (in Chinese). vol. 28, May. 2007, pp.2185–2188
[7]. H. Pang, S. Yang, L. Bian, “A Web Services Based Online Training and Exam System,” 4th
International Conference on Wireless Communications, Networking and Mobile Computing (WiCOM
'08), 12-14 Oct. 2008, Dalian, China, pp. 1 – 4
[8]. Z. Meng and J. Lu, “Implementing the Emerging Mobile Technologies in Facilitating Mobile Exam
System,” 2011 2nd International Conference on Networking and Information Technology IPCSIT
vol.17 (2011) © (2011) IACSIT Press, Singapore
[9]. Md. M. Islam, A. S. M. L. Hoque “Automated Essay Scoring Using Generalized Latent Semantic
Analysis” Journal of Computers, vol. 7, no. 3, March 2012.
[10]. W3SCHOOLS' ONLINE CERTIFICATION PROGRAM AT HTTP://WWW.W3SCHOOLS.COM/CERT/DEFAULT.ASP
[11]. L. Yang, G. Lian-sheng and W. Bin “Study and implement of distribution system based on J2EE and
MVC design pattern” Computer engineering and Design (in Chinese). vol. 28, Apr. 2007, pp.1655–
1658
[12]. K. Breithaupt, A. Ariel, and B. P. Veldkamp,“Automated simultaneous assembly for multi-stage
testing”. International Journal of Testing, 2005, page no. 319-330
[13]. H. S. Tan, S. Y. Hou, “Design and development of the universal testing system based on LAN”,
Journal of Central South University, 2000, page no. 367 – 370.
[14]. Z. Liu, L. N. Wang, and X. M. Dong, “Design and implementation of test paper reviewing system for
NIT network-based test”, Control Engineering of China, 2004, page no. 108- 110
[15]. Z. Zhang, G. H. Zhan, “The design and implementation of open and intelligent marking system for
computer operation test”, Computer Engineering and Application, 2001, page no. 14-16
[16]. Z. M. Yuan, L. Zhang, “G. H. Zhan, A novel web-based online examination system for computer
science education”, In proceeding of the 33rd Annual Frontiers in Education, 2003
Authors
S. M. Saniul Islam Sani has completed his B.Sc. degree from the Department of
Computer Science and Engineering from Chittagong University of Engineering &
Technology (CUET) in 2013. His research interest includes recommendation systems,
data mining, integrated software development.
Rezaul Karim is currently working as an Associate Professor, Department of CSE,
University of Chittagong. He is a PhD fellow at the Department of CSE, Chittagong
University of Engineering and Technology (CUET). His research interest includes data
privacy and data mining, data management, information retrieval, and social network
analysis.

22
Mohammad Shamsul Arefin received his B.Sc. Engineering in Computer Science
and Engineering from Khulna University, Khulna, Bangladesh in 2002, and
completed his M.Sc. Engineering in Computer Science and Engineering in 2008
from Bangladesh University of Engineering and Technology (BUET), Bangladesh.
He has completed Ph.D. from Hiroshima University, Japan. He is currently working
a Professor and Head in the Department of Computer Science and Engineering,
Chittagong University of Engineering and Technology, Chittagong, Bangladesh. His
research interest includes Data privacy and Data Mining, Cloud Computing, Big
Data, IT in Agriculture, and OO system development.

DEVELOPING A FRAMEWORK FOR ONLINE PRACTICE EXAMINATION AND AUTOMATED SCORE GENERATION

More Related Content

What's hot (20)

Similar to DEVELOPING A FRAMEWORK FOR ONLINE PRACTICE EXAMINATION AND AUTOMATED SCORE GENERATION (20)

More from AIRCC Publishing Corporation (20)

Recently uploaded (20)

DEVELOPING A FRAMEWORK FOR ONLINE PRACTICE EXAMINATION AND AUTOMATED SCORE GENERATION