An Learning Situation Early Warning Method Based on Linear Regression

Since the popularization of information technology, the teaching mode of college courses has been constantly innovated. The traditional offline course teaching mode is gradually replaced by online classes based on excellent open online courses, online and offline hybrid classes based on MOOC or SPOC or other online courses, and offline classes based on in-depth integration of information technology, this presents a new challenge for the early warning of learning situation in the course. For this reason, a new method is proposed for early warning of learning situation in the offline classroom which is deeply combined with information technology. This method takes offline classroom data and online platform data as teaching process data, normalizes each data item in the teaching process data by defining a series of models, and then constructs a calculation model of early warning value, which is used to dynamically monitor the change of early warning value of students in the process of classroom implementation. Once a student's early warning value exceeds the set value, the teacher can timely remind the student, find out the difficulties and reasons in the course learning, and help him return to the normal course learning as soon as possible, so as to gradually reduce his early warning value to a reasonable range. The comparison experiment proves that the early-warning value obtained by this method has a good linear regression relationship with the course examination results. The lower the early-warning value is, the higher the course performance is. This can be used by teachers as a reference to improve the quality of teaching, so as to encourage students to improve the quality of learning and improve the score of the course.