Published by : International Journal of Engineering Research & Technology (IJERT)

http://www.ijert.org ISSN: 2278-0181

Vol. 6 Issue 05, May - 2017

Evaluating Student’s Performance using

K-Means Clustering

Mr. Shashikant Pradip Borgavakar Mr. Amit Shrivastava

Research Scholar: C.S.E Asst. Professor: Computer Science & Engineering
Swami Vivekanand College of Engineering Swami Vivekanand College of Engineering
Indore, India Indore, India

This paper A fast and

Abstract— Data Clustering is the task of grouping a set of presented efficient
objects in such a way that objects in the same group are more iterative algorithm
similar to each other than to those in other groups. In this paper To
improved k- for data mining is
data clustering is used as k-means clustering to evaluate student means refining an that to
Specific Overcome
clustering initial discover the
performance. Evaluating student performance on basis of class research the problem
algorithm starting data and
test, mid test and final test. As we get cluster of student on this questions or of
that makes point for a patterns and
research Accuracy
basis of student marks will help to reduce ratio of fail student. the k-means general store it in an
objective and
more class of understandable
This information will help professor to student fail chance Efficiency
efficient and clustering form
before final exam.. produce algorithms
good quality has been
Keywords- k-means, Database, academic performance etc. clusters presented
Using K-
Means Iteration
INTRODUCTION Broad outline Applied
clustering improve k- Using
of how the DBSCAN and
Data clustering is a process of extracting previously Algorithm means Clustering
author solved OPTICS
and The cluster Cluster
unknown, valid, positional useful and hidden patterns from the problem algorithms
enhanced algorithm
large data sets (Connolly, 1999). The amount of data stored in Method
Phase 1 of
educational databases is increasing rapidly. Clustering the
technique is most widely used technique for future prediction. enhanced
algorithm
The main goal of clustering is to partition students into requires a
homogeneous groups according to their characteristics and time
abilities (Kifaya, 2009). These applications can help both complexity
of O(n2) for
instructor and student to enhance the education quality. This finding the
Dividing
number of
study makes use of cluster analysis to segment students into initial
parts
centroids,
groups according to their characteristics. Details of as the
thencalculate
Results on
All clustering
centres and algorithm
implementation maximum Real Word
decide process and
I. LITERATURE SURVEY of procedure time
memership
Data
find
required
Improving of patterns
here is for
the then rrepeat
computing
Accuracy Refining same steps
An Iterative Comparision the
and Initial distances
Improved k- of various
Research Paper Efficiency Points for between
means clustering
of the k- K-Means each data
Clustering algorithms
means Clustering point and
Clustering all other
Algorithm data-points
Estimate is in the set
fairly Key iterative
unstable define k
Lower Which contribution of improved k- K-Means
Number of due to centroids, Clustering
Problem being accuracy clustering the paper means clustering
Iterations are elements one for Clusters
addressed and algorithm is claimed by the clustering Algorithm
Less of the tails each cluster
efficiency best author. algorithm
appearing
in the
sample
Total II. DATA CLUSTERING
Importance
number of
algorithm iterations
of the Data Clustering is unsupervised and statistical data
problem of
Importance of requires a required by
having a
Way of analysis technique. It is used to classify the same data into a
the problem time k-means and Process
complexity improved k-
good homogeneous group. It is used to operate on a large data-set
initial
means is
points
to discover hidden pattern and relationship helps to make
much larger
Accuracy
decision quickly and efficiently. In a word, Cluster analysis is
and used to segment a large set of data into subsets called
Check To finding
Gap in the Efficiency
multiple Initial
Finding clusters. Each cluster is a collection of data objects that are
prior work is most algorithm
complicated
iterations Points similar to one another are placed within the same cluster but
to reducing are dissimilar to objects in other clusters.

IJERTV6IS050070 www.ijert.org 114

(This work is licensed under a Creative Commons Attribution 4.0 International License.)
Published by : International Journal of Engineering Research & Technology (IJERT)
http://www.ijert.org ISSN: 2278-0181
Vol. 6 Issue 05, May - 2017

III. CLUSTERING IN HIGHER EDUCATION VI. RESULT AND DISCUSSION

Education is an essential element for the progression and The model produced following results:
betterment of a country. Education makes a people perfect by
which he/she can participate in any progressive work for the Graph.1: Shows the relationship between GPA and
country. Education makes a country civilized and well- Attendance ratio.
mannered. Clustering in higher education means it classifies
the student by their academic performance. Lack of deep and
enough knowledge in higher educational system may prevent
system management to achieve quality objectives, data
clustering methodology can help bridging this knowledge
gaps in higher education system.

IV. PROPOSED MODEL

In university academic performance are measured by internal A. Data Arrangement in tables
and external assessment. Internal assessments are class test We grouped the students regarding their final grades in
marks, lab performance, assignment, quiz, attendance. several ways 3 of which are: • Assign possible labels that are
External assessments are previous semester grade and final same as number of possible grades. • Group the students in
semester grade. So, by taking the internal assessment and three classes “High” “Medium” and “Low”. • Categorized
previous exam grade and by using data clustering technique the students with one of two class labels “Passed” for grade
we can predict what will be the final grade of a student. above 2.20 and “Failed” for grade less than or equal to 2.20

1. If prev-grade=high, quiz=good,assignment=complete, Table 1

lab-performance=good ,class-test=good, Class GPA No of student Percentage
attendance=regular and then final-grade=good 1 2.00-2.20 2.20- 5 8.33
2. If prev-grade=average, quiz=good, 2 3.00 3.00-3.32 10 16.67
assignment=incomplete lab-performance=good Class- 3 3.32-3.56 3.56- 17 28.33
4 4.0 15 25
test=average and atendance=regular then final-grade= 5 13 21.67
average
3. If prev-grade=low, quiz=average,
assignment=incomplete, lab-performance= poor mid- Here, I cluster student among their GPA, that means, from
term=low and attendance=irregular then final- GPA 2.00- 2.20 we have 8.33% student. From 2.20-3.00
grade=low. student percentage is 16.67%. From 3.00-3.32 we have
The proposed model try to identify the weak students before 28.33%. From 3.32-3.56 percentage is 25% .The percentage
final exam in order to save them from serious harm. Teachers is 21.67% between GPA 3.56-4.00. The graphical
can take appropriate steps at right time to improve the representation of GPA and the percentage of student’s among
performance of student in final exam. the student is given below.

V. K-MEANS CLUSTERING ALGORITHM Graph 2: Number and percentage of students regarding to

K-means is an old and widely used technique in clustering GPA
method. Here, k-means is applied to the processed data to get
valuable information .The pseudo-code of k-means clustering
is given below.

Step 1: Accept the number of clusters to group data

into and the dataset to cluster as input values
Step 2: Initialize the first K clusters - Take first k
instances or - Take Random sampling of k elements
Step 3: Calculate the arithmetic means of each
cluster formed in the dataset.
Step 4: K-means assigns each record in the dataset
to only one of the initial clusters - Each record is assigned to
the nearest cluster using a measure of distance (e.g Euclidean
distance).
Step 5: K-means re-assigns each record in the
dataset to the most similar cluster and re-calculates the
arithmetic mean of all the clusters in the dataset.

Fig. 1 Generalized Pseudocode of Traditional k-means.

IJERTV6IS050070 www.ijert.org 115

(This work is licensed under a Creative Commons Attribution 4.0 International License.)
Published by : International Journal of Engineering Research & Technology (IJERT)
http://www.ijert.org ISSN: 2278-0181
Vol. 6 Issue 05, May - 2017

Class GPA No of Percentage

student
High >=3.50 28 46.67
Medium 2.20<=GPA<3.5 27 45
Low <=2.20 5 8.33

After clustering the student, we group the student into three

categories. One is High, second is Medium, and the last one
is Low. Graphical representation of these three categories is
given below:

Graph 3: Shows the percentage of students getting high,

medium and low GPA

REFERENCES
[1] Alaa el-Halees (2009) Mining Students Data to Analyze e-
Learning Behavior: A Case Study.
[2] Behrouz.et.al., (2003) Predicting Student Performance: An
Application of Data Mining Methods With The Educational
Web-Based System Lon-CAPA © 2003 IEEE, Boulder, CO.
[3] Connolly T., C. Begg and A. Strachan (1999) Database
Systems: A Practical Approach to Design, Implementation, and
Management (3rd Ed.). Harlow: Addison-Wesley.687
[4] Erdogan and Timor (2005) A data mining application in a
student database. Journal of Aeronautic and Space
Technologies July 2005 Volume 2 Number 2 (53-57)
[5] Galit.et.al (2007)Examining online learning processes based on
log files analysis: a case study. Research, Refelection and
Innovations in Integrating ICT in Education.
[6] Henrik (2001) Clustering as a Data Mining Method in a Web-
based System for Thoracic Surgery: © 2001
[7] Han,J. and Kamber, M., (2006) "Data Mining: Concepts and
Techniques", 2nd edition. The Morgan Kaufmann Series in
Data Management Systems, Jim Gray, Series Editor.
[8] Kifaya(2009) Mining student evaluation using associative
classification and clustering. Communications of the IBIMA
vol. 11 IISN 1943-7765.
[9] ZhaoHui. Maclennan.J, (2005). Data Mining with SQL Server
2005 Wihely Publishing, Inc

IJERTV6IS050070 www.ijert.org 116

(This work is licensed under a Creative Commons Attribution 4.0 International License.)