Scrutinizing Students Performance using Machine learning

With the huge demand for information technology, the Learning approaches of the students have constantly been evolved and innovated. Every student acquires a unique form of learning techniques and enforces their learnings on their ventures. The new ambience has expanded the demand for combining e-learning with classroom training. This combination of learning aspects has given rise to the integration of information technology and they provide huge challenges in scrutinizing the student's performance in any course. For this new ambience, a regression algorithm is proposed for scrutinizing student's performance. This algorithm explores both the behavioral and performance of the students in their tasks. They take the dominant attributes as learning progress data, normalize each data item in the learning progress data by identifying a set of models, and then build a measurement model of the student's early detection prophecy value when evaluating the student's learning ability. When the value is steep, professors keep track of the early detection prophecy value to observe student's challenges and explanations in course learning. This could help the students to scrutinize their performance in the course by deliberately reducing the early warning prophecy value. This algorithm shows that the prophecy value obtained by this process has a strong correlation with the course standardized test. The stronger the course results, the weaker the prophecy value. Both students and faculty will benefit from this to enhance their approach to the study the course.