Design And Implementation Of Comment Platform For Courses

With the huge number of online courses,it is of great significance for learners to find the most suitable course for them.There are also many resources for learners to choose from for the same technical explanation course,but it is difficult for learners to choose a suitable course from a large number of courses.Generally,there are two factors for choosing a suitable course:on the one hand,the user has the initiative and chooses the course through course information and comments;on the other hand,the platform has the initiative,and the platform recommends the suitable course to the user through rich information.Therefore,a third-party course review platform that digs into the use of reviews is very valuable.This paper proposes a cold start algorithm that uses graph convolutional neural networks to assist learning,on the basis of the traditional joint training methods,using the collaborative filtering representation obtained through the learning of the graph convolutional neural network to guide the learning of the joint training method,in this way,the learning results of the joint training method are improved.At the same time,a random probability strategy is introduced to prevent the joint training method from being affected by bad auxiliary information.Compared with the traditional joint learning method,the cold-start algorithm using graph convolutional neural network for auxiliary learning greatly reduces the problem of invalid learning.At the same time,through the random probability strategy,the model can be effectively prevented from being disturbed by poor quality auxiliary information during the learning process.Finally,experiments prove that in cold-start recommendation work,the cold-start algorithm using graph convolutional neural network for auxiliary learning has obvious advantages in recommendation results.This paper also combines theory and practice,using the graph convolutional neural network to assist learning cold start algorithm to develop the comment course platform and apply the algorithm to the personalized recommendation module.Through the system,users can view the course details,rich and valuable comment details,and visual chart data of each course.Beginners can assist their own stage learning through the learning route module of the platform.For the personalized recommendation of users,the user’s behavior data and learning route information are used,and the data of the entire platform is fully utilized to recommend the most suitable courses for users,thus providing users with a valuable third-party course review platform.This platform not only improves the credibility of the course comments,but also greatly improves the quality of the text content of the comments,and further processes the comment data in all aspects so that users can understand the overall evaluation of the course in a short time.