Using Classification Data Mining for Predicting Student Performance
Guna Sekhar Sajja, Harikumar Pallathadka, Khongdet Phasinam, Samrat Ray
Abstract
To remain competitive in today's competitive world, an institution must be able to anticipate student performance, classify people according to their talents, and motivate students to improve their performance on future tests. If you want to improve your academic performance, you should inform your pupils well in advance that they should concentrate their efforts on a certain subject. These kind of analyses assist institutions in lowering their failure rates. This research forecasts a student's success in a course using their previous performance in comparable courses. Using a variety of data mining methods, it is possible to uncover hidden patterns within massive data sets. These patterns may be highly valuable for analysis and predictions. Educational data mining is a term that refers to a collection of data mining applications that have been created for educational purposes. These systems examine the data of students and instructors in order to deliver relevant information. Analyses may be used to classify or predict. These include Random Forest, ID3, C4.5, and SVM, among others. The experimental investigation makes use of a student data set from UCI machinery.Keywords: Machine Learning, Data Mining, Classification, SVM, Student Performance