Download Available Online at www.ijetcr.org

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Document related concepts
no text concepts found
Transcript
International Journal of Engineering Research and Generic Science (IJERGS)
Available Online at www.ijergs.in
ISSN:2455-1597
Volume 1; Issue 1; Page No. 21-27
An Improved Analysis for E-Learning by Using Data Mining Process with a Simple Rule based
Classification Technology
Ganesh Kumar Jangid, Amar Singh Ratore, Hardik Kumar Patel
[email protected]
M.Tech. Scholar in Computer Science and Engineering
Ganpat University, Ganpat Vidyanagar, Mehsana-Gozaria Highway, Kherva, Gujarat, India
Abstract
E-learning environments are becoming increasingly popular in educational establishments. A popular form of e-learning is
online exams and quizzes. Most of the online exams and quizzes are inadequate as they fail to provide a thorough analysis
of the results due to which students are unacquainted of their weak areas in the respective subjects. Educators are finding
it harder to focus on students performing poorly in respective subjects and learners are finding it difficult to make a
decision about which learning materials best meet his/her situation due to more and more learning material available
online. This paper presents a design of a proposed system known as E-Tutor. The proposed system offers an e-learning
application integrated with an online quiz system which would predict the understanding level of students and
automatically recommend favourable learning content by analysing the performance of students in online quizzes using
classification algorithm in data mining.
Corresponding Author: Ganesh Kumar Jangid
Page
1. Introduction
Due to an incredible improvement in Internet knowledge and World Wide Web, the conception of online learning has
grow to be reasonably admired amid students and teachers all through because previous decade. As a consequence, online
learning systems are expansively used in schools and colleges incorporated with conventional classroom training to
encourage knowledge and enhanced considerate of concepts using audio/visual aids. These online learning systems
consist of audio & video instruction, tutorials, online books, text lessons, journals, short quizzes & exams. All these
stuffing are simply available to students on stipulate. Online quizzes and exams are one.
Such as trendy and imperative form of online learning apparatus. It provides an array of remuneration to both students as
well as teachers. Some of the recompense of online quizzes are as follows:
Reduced efforts
E-learning relieves the require for testing to be complete at a detailed hour, but it also makes testing a hassle-free task as
corrections are mechanical. In the cases of "Essay Question" tests, e-learning systems are regularly prepared with keyword
tracking tools that evaluation depending on what has been mentioned in the essays.
Uniqueness
Testing and quizzing can be completed inimitable by randomizing inquiry and answers picked from large pool of
questions slightly than just recycling through same sets of questions all over again
Instant Grading and feedback
Grading manually is possibly the majority time intense task for the mentor. Though, in online learning systems it’s just a
substance of few seconds. It’s wherever the trainers has the capability to observation on the strengths and weaknesses of a
learner and facilitate learning to essentially take consigning.
In-depth analysis
In-depth analysis can be accomplished on consequences to extract useful information regarding students.
Eco-friendly
Going from hard-copy tests/ quizzes to contribution the same capabilities online reduces utilization of goods such as
paper.
Powerful self-assessment tool
Testing and quizzing online will frequently supply the user with consequences immediately. This is excellent for students
since it allows them to distinguish what they did incorrect instantly, what they require to focus on, and how to progress
should they have to retake the test. However the quizzes obtainable online are insufficient as they fail to contribute a
complete analysis of consequences. Such inadequate quizzes only inform to student, the achieved and the questions which
went right/wrong. The assignment of the proposed system is to take existing system to the subsequently height. In
21
Keywords: E-learning, Data, Mining, Processing, Tools, FSS, SPA, UNIMAS
Ganesh Kumar Jangid, et al. International Journal of Engineering Research and Generic Science (IJERGS)
proposed system, the quiz segment is separated into dissimilar subjects and each subject is additional separated into 10
chapters. Each query in quiz has a text clarification, video explanation, indication link of the mainly appropriate study
substance on web and PDF content connected with it. The scores of a student for all the chapters of a individual issue will
be analysed and fed into categorization algorithm which will calculate presentation level of the student. If the presentation of
the student is dilapidated, then the respective faculty and student will be notified and recommendations for study substance will be
provided to student by system as well as faculty.
Abbreviations
PACR Performance Analyser and Content Recommender
C
Course
PL
Performance Level
2. Literature Survey
An environment study is complete to evaluation comparable existing systems used to execute student presentation
analysis. Three existing system are selected since these systems are comparable to the proposed system.
2.1 Faculty Support System
Shana and Venkatacalam has proposed a framework named Faculty Support System (FSS) which is low in cost as it uses
cost effective open source analysis software, WEKA to analyze the students’ presentation in a course available by
Coimbatore Institute of Technology of Anna University [1]. FSS is capable to analyze the students’ data animatedly as it
is capable to inform of students’ data animatedly with the stream of time to generate or add a new rule. The in form of
new rule is probable with the help from domain expert and the regulation is resolute by data mining technique such as
classification technique. Classification technique is used to calculate the students’ presentation. Besides, FSS focus on the
classification of factors that supply to presentation of students in a particular course.
2.2 Student Performance Analyzer
SPA is existing secure online web-based software that enables educators to view the students’ presentation and keep
pathway of the school’s data. The SPA is a tool planned for analyzing, displaying, storing, and getting feedback of student
evaluation data [3]. It is an influential analyzer tool used by schools worldwide to execute analysis and displays the
analysis data once untreated student data is uploaded to the system. The analysis is done by tracking the student or class to
get the overall arrangement of student or class. It helps to recognize the students’ presentation which is lower the
predictable level, at conventional level or above the expected level. This would allocate the educators or staffs to
recognize the current students’ presentation easily. Other than that, it enables a variety of kinds of students’ presentation
statement such as advancement report and accomplishment description to be generated.
2.3 Intelligent Mining and Decision Support System
In Minds helps University Malaysia Sarawak (UNIMAS) to monitor the presentation of a variety of areas in every
UNIMAS’s departments [2]. The system enables top and mid-management in UNIMAS to have a apparent look on the
areas that desirable concentration by looking at the statistics, revenues and risks. The features, ease of use and suppleness
provided by the system makes the presentation scrutiny in UNIMAS to be performed in an ideal solution. Charts are
provided by the system for ease of student performance’s explanation.
2.4 Student Analysis Performance system (SPAS)
Chew Li Sa, Dayang Hanani bt. Abang Ibrahim, Emmy Dahliana Hossain, and Mohammad bin Hossin have proposed a
system which offer a predictive system that is capable to calculate the students’ presentation in course “TMC1013 System
Analysis and Design”,. In University Malaysia Sarawak (UNIMAS) using BF Tree organization algorithm which predicts
whether the student is going to pass or fails in the current semester based on overall educational presentation during the
course of the whole semester.
A. Quiz
© IJERGS, All Rights Reserved.
Page
activities, this document uses simple regulation based categorization technique for classification activities.
3. Proposed System
Few descriptions from the existing systems are incorporated in the performance of the proposed system. These include the
easy user border, students’ performance prediction, diagram displays and report generation. In accumulation, the proposed
system consists of quiz segment and recommender which will suggest study substance to students. With all these
description in the proposed system, all the user requests will be contented. The proposed system is alienated into
subsequent modules/entities and these modules are included together to form a whole submission.
22
From the reviews on these existing systems, useful techniques and features could be functional into the
proposed system for an improved system’s presentation. Although, the above documents use WEKA for data mining
Ganesh Kumar Jangid, et al. International Journal of Engineering Research and Generic Science (IJERGS)
B. Performance Analyzer and Content Recommender (PACR)
C. Student
D. Faculty
Figure 1: Architecture of proposed system
3.1 Student Module
Student module includes student individual. All the students of an organization will have to register themselves on the
submission to admittance its features. Although register following attributes of student are in use into deliberation.
Attribute Name
Description
Student_ID
Id of student
Student Name
Name of Student
Roll_No
Roll Number
Class
Current Class of study
Division
Division
Email
Institutional Email
Contact
Contact Number
© IJERGS, All Rights Reserved.
Page
Registered Students can execute subsequent tasks
Online Quiz
Access the Online quiz by selecting respective centre and subdivision.
Track Progress
Student can verify their presentation chapter-wise/subject-wise and generally position amongst additional students simply
with help of diagram and tart charts.
Get Recommendations
If student is performing unfortunate in particular chapter/subject, he/she will be mechanically provided recommendations
for knowledge substance from system as well as individual subject’s faculty.
Help Section
Students can email their queries straight to respective subject’s faculty. Responses will be provided within 24 hours.
Student has to identify the subject name and subject followed by his uncertainty.
Feedback Section
23
Table 1: Student Registration Process
Ganesh Kumar Jangid, et al. International Journal of Engineering Research and Generic Science (IJERGS)
Student can present his precious reaction about the submission.
3.2 Faculty Module
Faculty element consists of faculty individual. The proposed system is faculty proscribed submission consequently faculty
has fundamental position in operational of appliance. Faculty has to register to work on appliance. Subsequent attributes
of faculty are in use into deliberation during register.
Attribute Name
Description
Faculty _ID
Id of Faculty
Faculty Name
Name of Faculty
Faculty_Subject
Subject which faculty Teaches
Email
Institutional Email
Contact
Contact Number
Table 1: Faculty Registration Process
Description
Question_ID
Specifies ID of Question
Course_Name
Name of Subject
Chapter_Name
Name of Chapter
Question_Text
Question
Question_Type
Specifies question Type Easy/Medium/Hard
First
First Option
Second
Second Option
Page
Attribute Name
24
Faculty performs following Task
Design Quiz
Faculties can add/modify/delete quiz questions.
Track Student Progress
Faculty can monitor presentation of students’ chapter-wise/subject-wise/class-wise easily with help of diagram and tart
charts.
Provide Recommendations
If student is performing arts poor in meticulous chapter/subject, respective subject’s faculty will be alerted so that faculty
can supply essential instructions and advise best study contented.
Solve queries Section
All of the student queries/worries can be answered in this section.
Feedback Section
Faculty can evaluate the feedback of students.
3.3 Quiz Module
Quiz module consist of multi-course quiz. Quiz section is separated into subjects and each subject consists of numerous
chapters. The Questions of the quiz are additional into the database by faculty only.
Creation of Quiz
Quiz questions are produced, deleted or customized by faculty members by faculty section.
© IJERGS, All Rights Reserved.
Ganesh Kumar Jangid, et al. International Journal of Engineering Research and Generic Science (IJERGS)
Third
Third Option
Fourth
Fourth Option
Final
Correct Option First/Second/Third/Fourth
Reference_Link
Link of the study material
File_Path
File path of the Study material Video
/Image/PDF
Table 3: Quiz database attributes
Working of Quiz
Quiz section is separated into Subjects and each theme consists of 10 chapters. Every episode has a group of simple,
medium and hard questions. Student has to choose subject and then section to start quiz consisting of 15 questions
containing easy, medium and hard questions (5 each) chosen in chance order. After submitting the quiz, a thorough report
will be provided to student consisting of subsequent
A. Score out of 15
B. Questions Correct & Wrong
C. Text Explanation of every question
D. Video explanation of every question
E. Reference link of online learning material
F. PDF/Image files
Quiz Scoring
Each quiz consists of 15 question containing 5 easy, 5 medium and 5 hard questions in accidental arrange. Easy
inquiry has 1 point, medium question has 2 points and hard questions have 3 points.
Difficulty
Points
Questions
Total Points
1
10
1 x 10 =10
Easy
10
2 x 10=20
Medium 2
3
10
3 x 10=30
Hard
Total Points for a quiz = 10 +20+ 30 = 60
Table 4: Total Score of each quiz system
Currently if a student get x easy, y medium and z hard questions correct out of 15 questions, then score in chapter i is
intended as:
………….… (1)
Student presentation level in chapter i is intended as:
………………………….. (2)
Page
categorization algorithm. If the presentation grade of the student is below AB then system generates an alert to
student as well as the individual subject’s faculty and recommends study substance to student stored in
database. The presentation level for all the chapters for which student takes quiz is intended. Lastly, the overall
presentation in a course i consisting n chapters is calculated as
25
3.4 Performance Analyzer and Content Recommender (PACR)
This component analyses the presentation level of student in all the 10 chapters of a subject and then classifies the
student’s presentation grade in that question as either AA, AB, BB, BC, CC, CD, DD using IF-ELSE rule based
© IJERGS, All Rights Reserved.
Ganesh Kumar Jangid, et al. International Journal of Engineering Research and Generic Science (IJERGS)
…………………….…. (3)
IF-ELSE rules are practical on subsequent quality
The classifier then assigns the final presentation grade of student based on Ci as:
Grade’s Value
Ci Value
AA
90 and above
AB
80-89
BB
70-79
BC
60-69
CC
50-59
CD
40-49
DD
Below 40
Table 5: Grades Classification Frequency
If the grade is below AB, then system mechanically displays the links of study materials connected with meticulous
subdivision in quiz database. Consequently, faculty can also propose student material through email facility.
4. Methodology Processing
During the expansion of project, subsequent phases existed.
© IJERGS, All Rights Reserved.
Page
4.1 Requirement gathering and analysis
The difficulty and data considerate is decisive in decisive the achievement of this submission .Before the system
expansion, problems and data sympathetic is recognized to describe the scheme objective and objectives. The problems of
the existing systems are recognized and analysed for its efficiency and competence in term of functionality. After the
problems are recognized, the solutions to resolve each difficulty is recognized and composed through journalism survey.
Moreover, other comparable systems are studied and analysed for its features, strengths and weaknesses. This helps to
recognize the needs and opportunities for the proposed system.
4.2 System analysis and design
In this phase, the generally stream of the system is designed, analysed and intended. The system and user necessities are
analysed. All the necessary scheme certification such as Software requirements specification (SRS), Project plan
certification and design certification is done. Additionally, all the design diagrams such as class diagram use case
26
Figure 2: Development phasing System
Ganesh Kumar Jangid, et al. International Journal of Engineering Research and Generic Science (IJERGS)
diagrams, data flow diagrams and ER diagrams are organized in IBM Rational Rose to capture significant aspects and
functionalities of the system. Intend of the proposed system also includes the design of database and user interface.
4.3 Implementation and testing
During the completion phase, the scores of the students in quizzes is composed and analysed by using easy regulation
base categorization algorithm in data mining to produce IF-THEN rules for classifying student into presentation grades.
These regulations will be implemented into the IF-ELSE circumstance using JAVA. This categorization assists the faculty
members to recognize those students who are performing poor in quizzes. Whole project is urbanized in eclipse IDE tool
and written in JSP. HTML, CSS, JavaScript is used for front end design and My SQL holds the database. request is hosted
on local host using Apache tomcat server.
Figure 2: Implementation of IF-THEN Rules
Page
27
4.4 Evaluation of System
For the assessment of the system, 20 end-users are requested to assess the usability of the submission. This is to ensure the
objectives of the proposed system are achieved as well as to ensure the ease of navigation across the interfaces of the
proposed system. Moreover, the assessment is performed to ensure the high efficiency of the proposed system is achieved.
From the evaluation performed, a list of users’ recommendation is confirmed as exposed lower
A. Including online tutorials and assignments in the application.
B. Appropriated recommendation generations.
5. Conclusion
The project focuses on the attractive of user learning knowledge through online quizzes. A data mining technique which is
a simple rule based classification algorithm is functional in this project to ensure precise calculation of the student’s
performance level in a subject by methodically analysing this quiz scores. The key payment of the proposed system is that
it assists the faculty in conducting student presentation analysis. The system assists faculty in identifying the students
whose presentation is poor. Furthermore, system mechanically recommends study material and even allows the faculty to
recommend study material to students having low performance taking the online quiz experience to a new level.
6. References
[1]. J.Shana, and T.Venkatacalam, “A framework for dynamic Faculty Support System to analyse student course data”,
International Journal of Emerging Technology and Advanced Engineering, Vol. 2, No. 7, 2012, pp.478-482.
[2]. Quality Assurance Division (2010). In Minds: Intelligent Mining and Decision Support System. University Malaysia
Sarawak, UNIMAS[Online].Available:http://www.qad.unimas.my/Function/ICTCompliance/inminds.html.
[3].SPA
(2013).What
is
SPA
Standard?
SPA
Student
Performance
Analyser
[Online].
Available:https://www.studentperformanceanalyser.com.au/spa/about.html.
[4]. Student Performance Analysis System (SPAS) Chew Li Sa, Dayang Hanani bt. Abang Ibrahim, Emmy Dahliana
Hossain, Mohammad bin Hossin Faculty of Computer Science and Information System University Malaysia Sarawak
(UNIMAS) 94300, Kota Samarahan, Sarawak, Malaysia.
[5]. Rule-Based Data Mining Methods for Classification Problems in Biomedical Domains Jinyan Li Limsoon Wong
Institute for Info omm Research h 21 Heng Mui Keng Terra e, Singapore 119613 Email: fjinyan, lampooni2r.a-star.edu.sg
July 26, 2004.
[6]. A Generic Framework for Rule-Based Classification Arnaud Giacometti, Eynollah Khanjari Miyaneh Patrick Marcel,
Arnaud Soulet L.I. University François Rabelais de Tours 41000 Blois, France.
© IJERGS, All Rights Reserved.