Sentiment analysis of MOOC reviews via ALBERT-BiLSTM model

The accurate exploration of the sentiment information in comments for Massive Open Online Courses (MOOC) courses plays an important role improving its curricular quality and promoting MOOC platform’s sustainable development. At present, most analyses are actually studies extensive sense, while relatively less attention is paid to such intensive issues as polysemous word familiar with upgraded significance, which results a low accuracy rate analysis model that used identify genuine tendency course comments. For this reason, paper proposed ALBERT-BiLSTM courses. Firstly, ALBERT was dynamically generate vectors. Secondly, contextual feature vectors were obtained through BiLSTM pre-sequence post-sequence, mechanism could calculate weight different words sentence applied together. Finally, output input into Softmax classification sentiments prediction sentimental tendency. experiment performed based on data set It proved result higher than already existing models.