Abstract
This paper designed a small course early warning system based on KNN algorithm, aimed at students haven’t finished the course of time can be completed by yourself some predict their courses by chance. In this paper, the basic principle of KNN algorithm is briefly introduced, and the course warning system is modified by Manhattan distance with added weights. This paper briefly describes the basic framework of this model and introduces the application of KNN algorithm in this model. Through a large number of basic experimental data to test the training, using figures to show, finally get the curriculum early warning system model, to achieve the effect of curriculum early warning.
Highlights
Before today's college students in the curriculum examination is only to pass a judgment about the course, this will lead to the students for the future study of tension, which affects the enthusiasm of students on this course
The course warning system discussed in this paper, with relatively small sample data and relatively monotonous sample attributes, perfectly makes up for its shortcomings
The above mentioned course warning system solves the problem that college students can only infer whether their courses can pass through by guessing an inaccurate and vague feeling
Summary
Before today's college students in the curriculum examination is only to pass a judgment about the course, this will lead to the students for the future study of tension, which affects the enthusiasm of students on this course. This course grades before the early warning system will be in the curriculum examination to a student's course can be through a more accurate prediction, in order to reassure students, make students actively to put energy into complete the requirements of this course. Starting with the introduction of KNN algorithm, the realization process of the early warning system grades is gradually introduced. Course grade warning system will past learning of this course students collected data as sample data, by through and not through into two categories, extracted from these two kinds of attributes related to whether through this course, students compare with the target, which can more accurately get the student through the conclusion of this course
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
