Year: 2024
Type: conference-paper
Abstract: This article takes the review data of a MOOC course as the research object, explores and analyzes the sentiment classification of Chinese short text data, and proposes an improved LSTM model to classify short texts. Firstly, the BERT pre-trained model is used to pseudo label the MOOC course review datasets based on different emotional features. Then, based on the LSTM model, a relational network is stacked downstream and attention mechanism is integrated. Finally, the text classification performance of the improved LSTM model will be experimentally compared with other text classification methods on a MOOC course review datasets. The experimental results show that all indicators of the model proposed in this paper have been improved, and the classification performance is the best.
Author: Liluxi Xu
Institution: Xi'an University of Architecture and Technology
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: gold
OpenAlex ID: W4409965501