Abstract
Learning management systems are mainly concerned to educational sectors play a dominant role in leading the nation across the globe. These days as everything goes online in terms of data storage across the globe. Lots of data emerges from learning systems and makes very promising to predict and classify learner’s performances. In order to classify students’ performance, various techniques are available. One of the most popular techniques to classify students’ performance is Machine Learning and is widely used in learning systems to process Informative facts about learners. Processing Educational data involves usage of several data processing methods like forecasting, clustering and finding out associations in order to extract the valuable information of the learners, their mood changes in shifting of subjects and accordingly their performances by extracting the hidden knowledge. Subsequently the obtained useful information and patterns can be used in predicting student’s performance. This research work suggests the effective technique in order to process and classify learner’s performance. Data is gathered from a middle east university concerning to graduate course. Machine Learning techniques like Support Vector Machine-SVM, Multi-Layer Perceptron-MLP, Random Forest-RF, Decision Tree-DT, Naïve Bayes-NB and K-Nearest Neighbor-KNN are applied after preprocessing the data. The outcomes attained are assessed on few metrics like Accuracy, Sensitivity, Specificity, ROC Curve Area and Kappa statistics. SVM outperforms in classifying learners’ part linked to other methods by yielding optimal classification results like high accuracy and Sensitivity followed by MLP, RF, DT, NB and KNN.

Key words: Educational Data; Support vector Machine (SVM); Multi-Layer Perceptron (MLP); Random Forest (RF); Decision Tree (DT); Naïve Bayes (NB); K-Nearest Neighbor (KNN).