Call us Today ! +918886268863 | [email protected]

Use of Classification Technique in Educational Data Mining

Introduction:

Data mining is used in various applications like banking, telecommunication industry, retail industry, DNA analysis, etc [2]. Data mining consist of six fundamental tasks: Anomaly detection, Association rule learning, Clustering, Classification, Regression, and Summarization. Educational data mining elucidate the application of data mining to the data generated from education setting. Ryan S. Baker and KalinaYacef [3] identified the following four goals of EDM:
● Predicting students’ future learning behavior
● Discovering or improving domain models
● Studying the effects of educational support
● Advancing scientific knowledge about learning and learners
There are four users involved in educational data mininglearner, educator, researcher, and administrator. There are various application in area of EDM such as- analysis of data, recommendation for the students, predicting the students’ performance, detecting the students’ behavior, etc. Here we considered how to use the classification algorithm to predict the class obtained by the students.

Abstract:
Data Mining is the process of identifying the hidden pattern from huge amount of data while Educational data mining is the technique used for automatically extracting the meaning from data that come from educational setting. Data mining algorithms are classification, clustering, association rule etc. Out of these algorithms, here the use of classification algorithm in Educational Data Mining for predicting the class (Distinction, First, Second, Pass, Fail, ATKT) obtained by the student in the university examination. In the current study, various classification algorithms results such as RandomForest, SimpleCart, DecisionTable, Naïve Bays, SMO, and SimpleLogistic are compared and finally the algorithm RandomForest is considered to explain the classification technique. Training set is used to train the algorithm for classification rule and test set is used to check whether classification rule correctly classified the test data.

Existing work:

Previous researches provided a comprehensive review of different classification techniques such as classification method including k-nearest neighbor classifier, Bayesian networks and decision tree induction in data mining.
Many researcher compare the various algorithms of classification technique in various field like crime and accidents [15], Cardiovascular Disease Prediction [16], breast cancer [17], Lung cancer [18] etc. Various researcher simply compares the classification algorithm based on various features or using different datasets.

Disadvantage:Many researcher compare the various algorithms of classification technique in various field but they not use of classification technique in Educational data mining.

Proposed work:

In this various classification algorithms are proposed such as Random Forest, Simple Cart, Decision Table, Naïve Bays, SMO, and Simple Logistic are compared and finally the algorithm Random Forest is considered to explain the classification technique. Training set is used to train the algorithm for classification rule and test set is used to check whether classification rule correctly classified the test data.

Advantages:
Usingdata mining tool Weka also matched with the real data of examination. And performance is high

Algorithms:
Random forest, Decision Tree, Naïve Bays

System requirements:
System requirements:
Software requirements:

• Operating system : Windows.
• Coding Language : Python.

Hardware components:
System : Pentium IV 2.4 GHz or intel
Hard Disk : 40 GB.
Floppy Drive : 1.44 Mb.
Mouse : Optical Mouse.
Ram : 512 Mb.

Conclusion:
In the current article, the classification technique is explained with the help of algorithm- RandomForest, in Educational Data Mining. This technique is used to predict the class (Distinction, First, Second, Pass, Fail, ATKT) obtained by the student in the university examination. Training set is used to train the algorithm and obtain the classification rules while test set is considered to predict the class using classification rules.

March 14, 2022

0 responses on "Use of Classification Technique in Educational Data Mining"

Leave a Message

Template Design © VibeThemes. All rights reserved.