Data Mining Techniques Based Students Achievements Analysis
Year: 2018
Type: article
Abstract: In this work, data mining techniques are used todetermine the students’ achievements in Mathematic class. In other words, weuse the data mining techniques to determine if there is any link between thestudent achievement and various student related data such as student grades,demographic, social information and school related data. Data miningtechniques, Decision Tree (DT), Discriminant Analysis (DA), Support VectorMachines (SVM), k-nearest neighbor (K-NN) and ensemble learner are used inprediction purposes. A publicly available dataset is considered in experimentalworks. Experimental works, on computer environment are carried out to validatethe data mining techniques. All data mining methodologies are simulated onMATLAB environment with 5-fold cross-validation technique. The classificationperformance is measured by accuracy and root mean square error (RMSE)criterions. Three experimental setups and for each setup, three scenarios areconsidered during experimentation. The obtained results are encouraging and thecomparison with some of the existing achievements shows the superiority of ourwork.
Authors: Dönüş Şengür, Songül Karabatak
Language: English
Subfield: Computer Science Applications
Field: Computer Science
Domain: Physical Sciences
SDG: Quality education
Open Access status: bronze
OpenAlex ID: W2939627690