Abstract
In recent years, sentiment analysis has made a great breakthrough in the field of natural language processing. Many researchers attempt to build a user comment analysis model of e-commerce products to infer the quality of goods. However, the application of sentiment analysis technology in online education platform and intelligent feedback to students and managers is almost irrelevant. In this paper, we collected a large number of online education platform web resource reviews to build a large corpus to pre-train the Word2vec model for comparative experiments. At the same time, the BERT+BiGRU model is designed for the Feedback System Structure (FSS), which achieves the effect of intelligent interaction between students and teachers and the improvement of the curriculum. On the self-built data set, the BERT+BiGRU model has the best prediction effect, with an accuracy of 98.82%. Compared with traditional machine learning methods such as Naive Bayes, the classification accuracy rate is improved by 21.54%. The experiment result shows that we use BERT+BiGRU on the FSS system to achieve the effect of teacher-student interaction and intelligent feedback of academic intelligence. In addition, we also carry out sentiment analysis on the text with Chinese Internet buzzwords. The experimental results show that the sentiment analysis model of our intelligent feedback system (FSS) has strong analytical capability.
Export citation and abstract BibTeX RIS
Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.