TASNIFLASH USULLARIDAN FOYDALANGAN HOLDA TALABALARNING O’QUV NATIJALARINI BASHORAT QILISH

Yuldasheva Saodat Chirchiq davlat pedagogika universiteti Pedagogika fakulteti katta o’qituvchisi

Abstract

Performance prediction has emerged as one of the most popular ways to use large amounts of online learning data. Most current work predicts performance based on students’ past performance on graded learning resources (e.g., problems and quizzes), while ignoring their performance on ungraded resources (e.g., reading materials). In this paper, we present an approach that can use student performance with non-assessed learning resources as supporting data to predict student performance on graded resources. This approach can reveal hidden relationships between different types of learning resources using only student activity data. Based on our experiments, the proposed approach can significantly reduce the error in predicting student performance compared to the original algorithms, while identifying meaningful and surprising relationships between learning resources.