Educational data mining in moodle data

<p>The main purpose of this research paper is to analyze the moodle data and identify most influencing features develop predictive model. The applies a wrapper-based feature selection method called Boruta for best predicting features. Data were collected from eighty-one students who enrolled in course Human Computer Interaction (COMP341), offered by Department Science Engineering at Kathmandu University, Nepal. University uses Moodle as an e-learning platform. dataset contained eight where Assignment.Click, Chat.Click, File.Click, Forum.Click, System.Click, Url.Click, Wiki.Click was used independent Grade dependent feature. Five classification algorithms such K Nearest Neighbour, Naïve Bayes, Support Vector Machine (SVM), Random Forest, CART decision tree applied data. finding shows that SVM has highest accuracy comparison other algorithms. It suggested File.Click System.Click significant This type helps early identification students’ performance. growing popularity teaching-learning process through online learning system attracted researchers work field Educational Mining (EDM). Varieties are generated several activities can be analyzed understand student’s performance which overall process. Academicians especially instructors use platforms delivery contents learners these highly benefited research.</p>