IPASJ International Journal of Information Technology (IIJIT)

Web Site: http://www.ipasj.org/IIJIT/IIJIT.htm

A Publisher for Research Motivation ........ Email:[email protected]
Volume 6, Issue 9, September 2018 ISSN 2321-5976

Road Map on Educational Data Mining

M. N. Quadri
Assistant Professor
Department of Computer Science
Nilkanthrao Shinde Science & Arts College, Bhadrawati,(M.S.) India

ABSTRACT
Educational Data mining is an important subject and research ground which are supposed taken care of the great data which
are arising in the educational field. The exploring data is formed as a data set and then various computational techniques are
applied to derive an interesting pattern out of it which further analysis, decisions can be done for the re-modeling of learning
pattern, improvement of teaching method and also for the improvement of education organization. This paper makes a wide
study about recent and relevant studies that has been gone through this area. The result of these studies will give the technical
process to taking up the teaching style into the future level and gives foreshadow the students’ educational performance.
Keywords: Data Mining, Education System, Educational Data Mining, Database.

1. INTRODUCTION
Educational Data mining is most demanding and innovative data mining sub domain in the research community.
Educational data mining is a research area which is a package of mathematical and dispositional method for
comprehending the learning process of students and also derives what are the strategies that should be taken up by the
teachers to enhance their teaching skills along with the improvement in the education system. Data is exploring
enormously in both traditional and web based educational system. To handle such a huge data exploration lots of tools
were used such as systematic learning approaches and education process management system. The success of the huge
work needs much more focused work in order of educational data mining [12] to become a strong and fully fledged
area.

2. EDUCATION AND SYSTEM

Education is a standardize process of providing or acquiring knowledge, skill, values, beliefs and habits [9]. Education
can be gained through any sort of methods like teaching, research, training, discussions, and brainstorming or even
through story telling. In all likelihood in the education process there will be a tutor to educate the learner, but it is not
necessary to have a tutor for acquiring education as the learner himself can gain knowledge through self-education.
Education can be classified as Formal Education and Informal Education. Formal education means the class room-
based education, which is provided to the learner by any type of educational institution with the help of a tutor or
trainer. In the other hand Informal education is a wide concept where the educational process took place outside of a
standardized classroom setting. Informal education is the sage, reverent and extempore process of learning. It works
through the constant interaction, exploration and augmentation of the learner’s experience. After the dawn of the last
decade of twentieth century the global education system has acquired a drastic change in its system. Across the borders
the education became as the most essential part of a human being in his career and personal life enrichment.
Organizations like UNESCO (United Nations Educational, Scientific and Cultural Organization), UNICEF (United
Nations Children Emergency Fund) [6] and various other social reforming teams took some solid step and a lot of
volunteer activities to give more inputs to the enrichment of quality and standard of education in global level. Later on
lots of developed and developing nations joined their hands together for the same cause of education development. They
also encourage to provide basic education to all the people who are not able to complete their primary education. RTE
which is an abbreviation of Right to Education is a prerogative to the development of education in International level
[11]. RTE imposes the freedom to acquire education beyond all discriminations and sets up the basic standard to

Volume 6, Issue 9, September 2018 Page 30

 IPASJ International Journal of Information Technology (IIJIT)
Web Site: http://www.ipasj.org/IIJIT/IIJIT.htm
A Publisher for Research Motivation ........ Email:[email protected]
Volume 6, Issue 9, September 2018 ISSN 2321-5976

enhance the educational system. The Education is isolated into three levels.
2.1 Primary Education
Primary education is compulsory and free of cost for all the children and not matter of their nationality or gender and
also there should not be any consideration for any sort of discriminations.
2.2 Secondary Education
It is also known as the elementary education and it should be generally available and accessible to all regardless to their
age.
2.3 Higher Education
It is also known as University level education and if a student or a learner is eligible to study a particular course by
meeting all eligibility criteria and university norms, then He / She is having the complete rights to acquire university
education.

3. STAKEHOLDERS OF EDUCATION SYSTEM

An individual, group of people or an organization those who are having interest or concern about an enterprise is
known as stakeholders. Educational stakeholders are the one who are directly depended to the educational system or the
one who are making favor to enhance the educational system [6]. The key stakeholders in educational system are
3.1 Teachers:
Teachers are the major stakeholder in education system and their main stake is to improve their professional efficiency
in the field of education along with the satisfaction which they are getting with this job.
3.2 Students:
Students are one of the key stakeholders in the education system. The primary intention of students are to acquire
knowledge that will help them to achieve success in their career and as well as in their personal life.
3.3 Parents:
Parents are also one of the stakeholders who feels gratification, triumph and chance for their wards whom they actually
aegis.
3.4 Institution and Government:
The stake of the institution and the government is to provide sufficient progress which is supposed to imply on the
education process of the institution for achieving the basic advancement in the system.
3.5 Corporate Business Community:
The corporate business community is also an educational stakeholder and their stakes are framed in such a process that
they will hire the graduate students who are having the desired skill set and required knowledge to work for their
business enterprise.

4. DATA MINING
Data Mining is a versatile sub domain of Computer Science and Technology. It is a process of finding similar or
interesting patterns from a large data set. Data Mining is a field where the user will come across Database systems,
Statistics, Artificial Intelligence and Machine learning too. The major concept behind to the data mining is to process
and extract information or data from a large and data set and convert it into understandable structure for the future use
[8].
4.1 Knowledge Discovery of Data (KDD)
Knowledge Discovery of Data, an acronym of datamining is the process of extracting hidden and interesting knowledge
from data[1].

Volume 6, Issue 9, September 2018 Page 31

 IPASJ International Journal of Information Technology (IIJIT)
Web Site: http://www.ipasj.org/IIJIT/IIJIT.htm
A Publisher for Research Motivation ........ Email:[email protected]
Volume 6, Issue 9, September 2018 ISSN 2321-5976

The steps of KDD process is as follows,

Problem Study: Acquire prior knowledge about the application domain and understand the goal of the KDD process.
Creation of Target Data: Selection of data of interest from the data set.
Data Cleaning: Making of clean data by removing the noisy,redudant and irrelevant data.
Data reduction and transformation: Reduction of data set by choosing dimensions of interest and transforming to the
format that can be interepted by the mining methods.
Applicaion of Data Mining Methods: Based on the goal of the KDD process any of the mining intelligent methods
like summarization,clustering,classification,regression and so on can be applied accordingly.The resultant of this
application is data patterns .
Evaluation of patterns: Data patterns acquired are evaluated for interestingness.
Knowledge presentation: Interesting patterns are interpreted and presented as models for visualization of knowledge.

Preprocessed Transformed
Data Data

Target Data Model

Data Knowledge

Figure 1 Process of Knowledge Discovery of Data

a. Data Mining Methods
Major Data mining techniques [1] are as follows
Association : This technique is also called as relation technique. The pattern discovered is the relationship between the
items in the transactions.
Classification : It classifies the item of the dataset into predefined classes or concepts.Mathematical techniques like
decision tree,neural network and so on is utilized to classify the item.
Clustering : Organize the items of similar characteristics as clusters whose class labels are unknown.
Regression : Discovers the relationship between the variables. The relationship may be between independent variables
and dependent variables.
Sequential Patterns : This analysis discover the regular and frequent patterns in the data set.

5. DATA MINING IN EDUCATION

Educational data mining (EDM) [4] is a blooming discipline that develop methods to explore the large and wide variety
of data that are obtained from educational institutions and to use those data to find out the educational style and
learning environment of the institution and also to understand the character and learning style of students along with
the teaching pattern of the faculties.

Volume 6, Issue 9, September 2018 Page 32

 IPASJ International Journal of Information Technology (IIJIT)
Web Site: http://www.ipasj.org/IIJIT/IIJIT.htm
A Publisher for Research Motivation ........ Email:[email protected]
Volume 6, Issue 9, September 2018 ISSN 2321-5976

Figure 2 Educational Data Mining

During the past few years the most innovative things which happened in the field of education is the implementation of
new technologies [12]. Class room based traditional education has grown up to be a web based education which can be
done with a computer system with network connectivity and it is virtually independent of any location or hardware
resources. Growth of web based education was tremendous in last couple of years. In traditional teaching environments,
tutors or instructors are able to teach the learner with a face to face interaction and also they are able to make
continuous assessment regarding their learning programs. In the traditional class room system decision making can be
done by observing student behavior, conducting continuous real time examinations and also by estimating the training
effectiveness. But all these leverages of assessment can’t be done with web based learning. Trainers are supposed to
find out some other ways to gain all these information. Online education sites collects large amount data with various
automated cloud based techniques with the help of web servers and collected in server access logs. Such web based
learning are able to collect wide range of learning habit of students along with the data consist of their general interest
and hence very large amount data is produced. Currently the demand for web based education is increasing rapidly so
in order to provide more quality and productivity in the education system data mining technologies can be
implemented.
5.1 Phases of Educational Data mining
As emerging research area educational data mining, multitude of data mining techniques is applied to a variety of
educational contexts. The aim is to transform data into information that helps to make better decision making on design
and path of a academic environment to enhance the learning process. EDM consists of four phases [12],
Applying Data
Mining Techniques Future
Prediction

Data Decision
Collection Making

Figure 3 Educational Data Mining Phases

5.1.1 Data Collection: Target data is collected from various educational system repositories and preprocessed.
5.1.2 Association Discovery: This phase aims at discovering associations between the data variables. Various
algorithms like classification, regression, association rule mining, clustering, sequential patterns, and decision tree
and so on are applied to discover the relationships.
5.1.3 Validation of Association: To avoid over fitting, the identified associations should be validated.
5.1.4 Decision Making: The validated associations are utilized for predictions which support the decision making to
enhance the learning environment.
The Data is visualized in the phase 3 and 4 to avoid knowledge dilution.

Volume 6, Issue 9, September 2018 Page 33

 IPASJ International Journal of Information Technology (IIJIT)
Web Site: http://www.ipasj.org/IIJIT/IIJIT.htm
A Publisher for Research Motivation ........ Email:[email protected]
Volume 6, Issue 9, September 2018 ISSN 2321-5976

5.2 Outcomes Of Educational Data Mining

Educational Data mining benefits stakeholders by the following resultants [10],
Student Modeling: Student modeling in educational field subsumes variety of information such as student’s
attributes like comprehension to grasp an information, understanding own thought process, competence, impulse,
gratification, attitude or some sort of learning process which made a negative impact with its ultimate learning
outcome. The generalized objective improve student model.
Predictive Modeling: The main objective is to predict a student’s overall performance or learning outcomes
regarding [3] to the course activity data.
Generating Recommendations: The objective is to recommend the students about the task which are most
appropriate for them.
Analyzing learner’s behavior: With help of various forms, analyzing the learner’s behavior by applying various
data mining techniques.
Maintaining and improving courses: The objective here is to predict how to improve the learning activities of the
students using various information available and also to identify the teaching sequences which can be adapted to
improve the student’s need.
Learners: It is to provide recommendations to the learners and also provide appropriate feedbacks to student’s needs
which will improve their learning performance.
Educators: To improve their own teaching method by incorporating the learning process and behaviors.
Administrators: To assess the institution’s infrastructure, resources and materials to improve the quality of the
educational system.
5.3 Research on Educational Data mining
Some of the research studies in the emerging research domain educational data mining is as follows [5][3],
Student Modeling Research: Student modeling is one of the most popular research areas in the education data
mining field. The scope of student modeling ranges from unmanned improvement of student model to Coalesce
discovery in student understanding model.
Predicting Future Learning Better Using Quantitative Analysis of Moment-by-Moment Learning: It is mainly
focused on predicting how a student performs in his future based on the skills learned and also predicting how the
student has been prepared himself for future learning.
Discovering Student Models with a Clustering Algorithm Using Problem Content: To overcome the traditional
way of making the student models.
Automated Student Model Improvement: For investigating the most inaccurate part of a model which helps to
derive new perception in the learning style of a student and also to recommend specific development tutor design.
Student Profiling from Tutoring System Log Data: Generating faults, inkling identification of students are
captured for portraying their learning strategies. More efficient for web-based education system.
Automated Discovery of Speech act Categories in educational games: Completely based on voice based clustering
methods
Sequences of Frustration and Confusion, and Learning: It is to identify the occurrence and impact of confusion
and irritation which affects a learning in a particular duration of time during their online web-based learning and it
has been identified that occurrence irritation is having more impact in learning process rather than of confusion.
Predicting drop out from social behavior of students: Assessment of social behavior to improve the accuracy of the
prediction.
A Spectral learning approach to knowledge tracing: It is used to learn PSR that resembles the Bayesian
Knowledge Tracing.
Mining Meaningful Patterns from Students Handwritten Coursework: This technique is applied to get the
students hand writing in a digital format and data mining techniques are implemented on the digital copies to
identify the mental process of a student during the normal environment.

Volume 6, Issue 9, September 2018 Page 34

 IPASJ International Journal of Information Technology (IIJIT)
Web Site: http://www.ipasj.org/IIJIT/IIJIT.htm
A Publisher for Research Motivation ........ Email:[email protected]
Volume 6, Issue 9, September 2018 ISSN 2321-5976

6. CONCLUSION
Educational data mining is one of the most challenging data mining research areas where still huge numbers of
innovative concept are flourishing up for mining new things from the existing dataset. Education can be delivered in
any form, because sharing or teaching someone about a specific or generalized subject can be characterized under
education system and not only the intellectual thoughts but also it will focus up creating a student or a group of persons.
Thus information or data set related to educational data is generating in a very large amount and data mining helps to
find out the most interesting pattern from such datasets.

REFERENCES
[1] Amala Jayanthi.M*, Swathi.S and Tharakai.R, “ Data Mining– A Survey”, International Journal of Advanced
Research in Computer Science and Software Engineering , Vol 6, No. 4, April 2016
[2] Brijesh Kumar Baradwaj and Saurabh Pal, “ Mining Educational Data to Analyze Students’ Performance ,”
(IJACSA) International Journal of Advanced Computer Science and Applications, Vol. 2, No. 6, 2011 .
[3] C.Romero and S.Ventura, “Educational Data Mining.A Survey from 1995 to 2005,” Elesvier,Science Direct,Expert
Sysytems with Applications, vol. 33, pp. 135-146, 2007.
[4] Crist´obal Romero, “Educational Data Mining: A Review of the State of the Art” VOL. 40.`
[5] D.Fatima, Dr.Sameen Fatima and Dr.A.V.Krishna Prasad, “A Survey on Research work in Educational Data
Mining”, IOSR Journal of Computer Engineering (IOSR-JCE) e-ISSN: 2278-0661,p-ISSN: 2278-8727, Vol. 17,
No. 2, , PP 43-49, Mar – Apr. 2015.
[6] Engaging Stakeholders”, Sustaining Reading First ,Sustainability Series Number 6 • September 2009.
[7] Geeta Kashyap and Ekta Chauhan, “Review On Educational Data Mining Techniques” ,International Journal of
Advance Technology in Engineering and Science,Vol No.3,Issue 11,November 2015.
[8] Hardeep Kaur, “ A Review of Applications of Data Mining in the Field of Education ”, International Journal of
Advanced Research in Computer and Communication Engineering, ISSN (Online) 2278-1021, Vol. 4, No.4, pp.
2319-5940, April 2015.
[9] http://knowledgeworks.org/worldoflerning/2015/07/studentperspective –future-learning/
[10] https://en.wikipedia.org/wiki/Educational_data_mining
[11] https://en.wikipedia.org/wiki/Right_to_education
[12] Rajni Jindal and Malaya Dutta Borah, “A Survey On Educational Data Mining And Research Trends,”
International Journal of Database Management Systems ( IJDMS ) Vol.5, No.3, June 2013, DOI
:10.5121/ijdms.2013.5304 53.

Volume 6, Issue 9, September 2018 Page 35