=Paper=
{{Paper
|id=Vol-1819/edudm2017-paper1
|storemode=property
|title=
|pdfUrl=https://ceur-ws.org/Vol-1819/edudm2017-paper1.pdf
|volume=Vol-1819
|authors=Anamika Jain,Arpana Rawal
|dblpUrl=https://dblp.org/rec/conf/indiaSE/JainR17
}}
====
https://ceur-ws.org/Vol-1819/edudm2017-paper1.pdf
BUILDING SUBJECT PRE-REQUISITE COURSE MAPS
FOR ON-GOING ACADEMIC COURSES
Anamika Jain Dr. Arpana Rawal
Research Scholar Professor
Bhilai Institute of Technology, Durg Bhilai Institute of Technology, Durg
Chhattisgarh, India Chhattisgarh, India
anamika20001@rediffmail.com arpana.rawal@gmail.com
ABSTRACT stretching ends constrained from contemporary curriculum
Educational data mining is an emerging discipline, concerned development, quality assurance, accreditation schemes, policy
with developing methods for exploring the vast data sets that planning and Governance ethics. Nevertheless, gradual but
come from live educational settings in learning environment.. It is consistent schematic amendments are taking place in order to
a continual process, keeping pace of vision and mission of an make the students’ put their academic efforts at a regular learning
institution, also addressing to issues in an innovative , ethical and pace. Academic analytics is an upcoming machine learning
responsive manner to meet the academic and administrative paradigm that was introduced to suffice such an objective of
objectives. Invariably EDM studies orient towards a mandatory enhancing higher educational quality standards. One way to
step of undertaking a cross sectional view of attributes achieve quality objectives in higher education system is by
contributing to learning patterns of the students as a whole. In this discovering analytical knowledge from educational datasets to
paper, a novel method is proposed to build subject pre-requisite study the major contributing attributes that affect the students’
course maps of higher semesters for students pursuing bachelors performance directly or indirectly. If Academic communities are
of engineering courses. The dependency computations are done able to identify the weak performers and slow learners much
by analyzing their performance in subject prerequisites of earlier to their duration of examinations using EDM practices, this
previous semesters. This piece of work focuses on entirely new knowledge can aid them in taking pro-active actions, so as to
kind of feature vector that is heavily subsumed to affect the most reframe better educational policies and strategies for enhancing
critical mining objective that is predicting subject wise students academic performance of these students with their upgraded
academic performance well before they face their end semester evaluation settings.
examinations.
2. ARM IN EDM PRACTICES
CCS Concepts A huge span of time has already been spent by EDM
• Information systems➝Database management system researchers revealing students’ profile patterns and understanding
engines • Computing methodologies➝Massively parallel and students’ learning behaviour using predictive modeling methods
high-performance simulations. This is just an example, please to identify drop out students, all encompassed together in a field
use the correct category and subject descriptors for your of software development called Academic Analytics. The realm
submission. The ACM Computing Classification Scheme: combines technology, information retrieval, management of data,
statistical analysis to tap these potential patterns that help faculty
http://www.acm.org/about/class/class/2012. Please read the HOW and advisors to become more proactive in identifying at-risk
TO CLASSIFY WORKS USING ACM'S COMPUTING students and responding to them for counseling activities and
CLASSIFICATION SYSTEM for instructions on how to classify subsequent remedial actions accordingly. Relationship mining has
your document using the 2012 ACM Computing Classification historically been the most exploited field of EDM research and
System and insert the index terms into your Microsoft Word remains extremely prominent to this day. Relationship mining is
source file. used to find the relationships between the created model’s
predictions and additional variable. These set of techniques enable
Keywords researchers to formulate association rules, correlation rules,
Educational Data Mining, Association Rule Mining, Subject Pre- sequential / temporal association rules and rules of causality.
requisites, Strong Association Rules, Rule Generation Association Rule Mining (ARM) is one of the most important and
well surveyed mining techniques intended to identify strong rules
discovered in databases using different measures of
1. INTRODUCTION interestingness. It aims to extract interesting correlations, frequent
Over years, higher education system in India has taken the shape patterns, associations or casual structures among sets of items in
of business oriented enterprise-cum-knowledge imparting the transaction databases or other data repositories [1] [2] [3].
portfolio gaining a fierce competitive edge at the end of
stakeholders and investors. The education system in our country
encounters enormous, diversified & manifold challenges with Towards the satisfaction of teaching learning objective in
academia, huge span of time has already been spent analyzing
Copyright © 2017 for the individual papers by the papers’ authors.
Copying permitted for private and academic purposes. This volume is varied students’ profile patterns; however very little effort is put
published and copyrighted by its editors. up for identifying subject wise learning levels of students by
exploring wide spectrum of attributes they exhibit while studying
their ongoing courses (subjects). With an exhaustive survey, a (Candidate Subjects) subject pre-requisites
general opinion upon which the entire EDM community is
unanimously agreeable that it is always better to undertake B.E. (Eighth semester)
rational educational parameters of the present rather than past to Artificial Intelligence (AI) NA, DIS, CF, CM,
construct predictive modeling of various forms [3] [4] [5] [6]. OOPS, M3, PSLB,
DBMS
3. PROBLEM HYPOTHESIS
For a given courseware, if the syllabus in any academic B.E. (Sixth semester)
curriculum provides a glimpse of the course map defining subject
pre-requisites for the subjects in forthcoming semesters, this may Computer Graphics (CG) M3, ADA, DS,
increase a concern of interest, inquisitiveness and sincerity among CSA, CF, DIS
these grooming minds towards the on-going courses. Verbal Owing to fact that in some of the configuration settings, too many
orientation and briefing sessions can be given on these course or almost nil association rules were extracted, the values of
maps in early days at the commencement of on-going academic support = (0.3) and confidence = (0.6) were finally selected as
semesters. A set of subject pre-requisites for any subject S is optimal values for extraction of interesting association rules for
defined as a set: S1,S2,S3……Sk, that the student studies in the every subject mentioned in the table 1, as a consequence of
curriculum of previous semesters and helpful for concept building multiple execution trials with varying combinations of
and thorough learning of the subsequent subject, ‘S’ in current configuration setting combinations. One such set of interesting
ongoing semester. Another way of interpreting the above problem correlations for B.E. 8th semester subject, ‘Artificial Intelligence
formulation is that, upon analyzing the scoring patterns of passed- & Expert Systems’ is shown in table 2 along with confidence
out batches in hypothesized set of subject pre-requisites, one can measures.
reveal strong co-relations between the subjects of on-going
courses and some of the selected pre-requisites called as strong Table 2 Strong association rules for Artificial Intelligence &
subject pre-requisites. Such computations are performed in the Expert Systems
following section using Association rule mining method. 1 CM=1 OOPS=1 134 ==> AIES=1 97
4. DATA SETS AND EXPERIMENTATION conf:(0.72)
4.1 Data Collection 2 NA=1 CF=1 144 ==> AIES=1 97
The students’ data sets collected in the current research study conf:(0.67)
pertain to the different courses (subjects) pursued by the students 3 CF=1 OOPS=1 153 ==> AIES=1 101
of engineering final and pre-final years. Students’ performances in conf:(0.66)
respective subject pre-requisites were collected from the
departmental records of result summaries pertaining to three 4 OOPS=1 184 ==> AIES=1 121 conf:(0.66)
passed out batches of computer science and engineering
5 M3=1 CF=1 151 ==> AIES=1 99
discipline. In the current study, end semester portion of the subject
conf:(0.66)
scores obtained, out of ‘eighty’ were recorded as separate input
attributes to the experimental setup. 6 CF=1 CM=1 151 ==> AIES=1 99
In the initial stages of experimentation, the subjects that were conf:(0.66)
perceived to be more or less influential for smooth grasping of a 7 A=1 182 ==> AIES=1 117 conf:(0.64)
subject in higher semesters of the mentioned discipline were
included as participating candidates for hypothesizing their pre- 8 PSLB=1 CF=1 154 ==> AIES=1 99
requisite schemas. Table 1 shows the pre-requisite schemas for conf:(0.64)
two subjects of sixth and eighth semesters, hypothesized after
9 CM=1 DIS=1 152 ==> AIES=1 97
taking expert opinions by eminent subject experts, who are also
conf:(0.64)
the nominated Board of Studies members in the affiliated
University, for the formulation of revised syllabi and schemes for
courses of these semesters. The abbreviations used for each
subject pre-requisite have their interpretation, the annotations Table 3 Dependency rule generation for subject pre-requisites
borrowed from the syllabi of almost all Indian Universities. of AIES and CG courses
Candidate Subject Rule Interpretation using
The experimental setup begins by fetching the ‘marks obtained’
subject pre-requisites
data columns by the passed-out student instances in the above
mentioned subject pre-requisites. Student’s scores in these Artificial Intelligence and IF (NA=1) or (OOPS=1)
prerequisites various subject pre-requisites of previous semesters Expert Systems (AIES) THEN AI=1
were analyzed and converted to nominal values in the form of 0 &
1. The WEKA ARM classifier begins with configuring the tool (8th Semester) ELSE IF (CM=1 and
option using suitably chosen values of ‘support’ (τ) and OOPS=1) or
‘confidence’ (σ) mining parameters. (NA=1 and CF=1) or
Table 1 Recommended subject pre-requisites for sixth and (CF=1 and OOPS=1) or
eighth ongoing academic semesters in Computer Science and
engineering discipline (M3=1 and CF=1) or
Subjects under Experiments Hypothesized (CF=1 and CM=1)or
(PSLB=1 and CF=1) or The rule interpretation task for two subjects, each from sixth and
eighth semesters is illustrated in table 3.
(CM=1 and DIS=1) or
(CF=1 AND DBMS=1)
THEN AI=1
Subject pre-requisite course maps identified for all semesters in
ELSE AI=0 the any academic curriculum is bound to benefit all learners and
teachers in teaching-learning environments. The findings of this
study conclude that some subjects act as heavily determining
Computer Graphics IF (M3=1 or DS=1 or criteria towards the prediction of student performance in higher
CF=1 or CSA=1 or ADA=1) semester subjects. Dependency structure thus obtained in the form
(CG)
of subject pre-requisite course map can be further exploited in
THEN CG=1
(6th Semester) prediction of subject wise scores of each student as a mark of
ELSE progressive evaluation framework, well before they face their
final examinations of the on-going semester. Identifying the
IF(CSA=1 and ADA=1)or subject attributes, that contributes the most towards the students’
(DS=1 and ADA=1)or performance help to improve the support services for students
who perform poorly in their studies in early stages amidst their
(M3=1 and ADA=1)or academic semester. It may be noted that similar experiments can
(M3=1 and DIS=1)or be extended upon total scores of the subject pre-requisites,
following the cumulative marking scheme based on internal
(DIS=1 and assessments too.
ADA=1)THEN CG=1
5. ACKNOWLEDGMENTS
ELSE CG=0 This work was supported by Research and Development
Laboratory, Department of Computer Science and Engineering at
Bhilai Institute of Technology, Durg, Chhattisgarh, India,
In the next step of computation, the strong association rules for awaiting sponsorship from suitable funding agencies.
the target (candidate) subject (with confidence value above
threshold) were exploited for interpreted rule interpretation task in 6. REFERENCES
the form of IF-THEN statements holding compound conditional [1] Bhardwaj B., Pal S.(2011), “Mining Education data to
expressions as antecedent stubs evaluating to strong dependency Analyze Students’ Performance”, International Journal of
of the candidate subject on the specified pre-requisites using Advanced Computer Science and Applications, Vol. 2,No 6,
following heuristics: 63-69.
1. Upon analyzing each rule it was observed that some [2] Kumar V., Chadha A.(2012), “Mining Association Rules in
pre-requisites contributed towards the dependency of candidate Student’s Assessment Data” , International Journal of
subject (consequent) independently. For instance, rules 2,4,7,8 Computer Science issue, Vol. 9,NO. 3, 211-216.
and 9 contribute the dependency to candidate subject ‘CG’ [3] Angeline D.M.D.( 2013), “Association Rule Generation for
independently and so appear as disjunctive components of the Student Performance Analysis using Apriori Algorithm”,
compound conditional expression in the antecedent portion of The Standard International Journal on Computer Science
rule. Engineering & its Applications, Vol. 1, No. 1, 12-16.
2. On the other hand multiple pre-requisites appear as [4] Singh M., Singh J., Rawal A.(2014), “Feature Extraction
conjunctive components and at least one occurrence out of many Model to identify At-Risk Level of students in academia”,
such combinations may contribute to strong dependency of the Information Technology (ICIT),2014 international
subject candidate; all together combined using disjunction conference on 22-24 Dec 2014, 221-227, ISBN-978-1-4799-
connectives. 8083-3.
[5] Singh M., Singh J.( 2013), “Machine Learning Techniques
for prediction of subject scores: A Comparative Study”,
IJCSN, Volume 2,Issue 4, August 2013, ISSN- 2277-5420.
[6] Singh M., Rawal A., Dubey S. (2010),”Modeling C4.5
Figure 1 Subject pre requisite course map for sixth and eighth Decision Tree Classifier for prediction of subject scores in
semesters ongoing courses of Academia”, NCICT 2010.