Intelligent Analysis Algorithm of Online Collaborative Learning Behavior of Chinese-Foreign Cooperative Students Combining Deep Learning and Transfer Learning

With the rapid development of online education, behavioral intelligence analysis of Chinese-foreign cooperative students in collaborative learning environment has become one of the important research directions in the field of education. This study proposes an algorithm combining deep learning (DL) and transfer learning (TL), aiming at effectively identifying and analyzing the online collaborative learning behavior of Chinese and foreign cooperative students, and providing data support and decision-making reference for educational practice. Firstly, the paper introduces the basic principles and common methods of DL and TL, and discusses their applications in intelligent analysis of online collaborative learning behavior in detail. Then, the paper designs and implements an algorithm model that comprehensively utilizes DL and TL, and combines the convolutional neural network (CNN) technology to capture students' behavior patterns and knowledge level. Then, the effectiveness and superiority of the proposed algorithm are verified by experimental design and result analysis. The experimental results show that our algorithm has achieved higher accuracy and recall rate in identifying students' learning status and knowledge level, and has significant advantages compared with the single method. In addition, the paper further analyzes the generalization ability of the algorithm in different countries and course types, and further improves the performance of the algorithm through parameter tuning. On the whole, the algorithm combining DL and TL proposed in this study provides an effective solution for the intelligent analysis of online collaborative learning behavior of Chinese and foreign cooperative students, which has important theoretical and practical significance.