Sentiment Analysis of MOOC Reviews Based On Capsule Network

With the development of Internet technology, the mode of "education + technology" is becoming more and more popular. More and more users begin to use MOOC platforms to select the courses they are interested in and leave rich comments with emotional colors in the comment area. The exploding number of comments contains information about learners' attitudes and evaluations of the course.It is very important to analyze and process these comments so as to obtain the key information.In this paper, a comment text sentiment analysis model ALBERT-BiGRU-Capsnet is proposed based on Albert pretraining language model, which integrates BiGRU and Capsule Network.Through many comparative experiments, it is found that the classification effect of this model is better.