Empirical Research on the Effect of Collaborative Learning in Blended Learning Mode Based on KNN Algorithm

Abstract

The quality of collaborative learning is one of the essential factors that determine the quality of teaching. Therefore, it is a significant work for educators to explore scientific and reasonable grouping methods. In this paper, first we design a Blended Learning mode in which there are a variety of online and offline learning activities. The quantified learning behavior information becomes the original data and basis for grouping. Then we combined KNN (k-Nearest Neighbor) algorithm and grouping principle to implement grouping for the pilot class. Finally, the effect of this grouping method is demonstrated by comparing the final examination results and analyzing the number of students who have finished the preview. The results show that the class with the new grouping method has achieved good performance in the final examination.