Addressing Sentiment Classification in Short Text Comments Using BERT and LSTM

DOI

10.2991/978-94-6463-540-9_83How to use a DOI?

Keywords

Natural Language Processing; Sentiment Classification; BERT; LSTM

Abstract

The prevalence of short text comments in the comment sections of social media platforms accelerates the rate of information dissemination. The diversity and unpredictability of comment content can affect the sentiments of viewers and their judgment of topics and interfere with social media platforms’ control over hot topics and optimization of user experience. Prioritizing the content of the comments section and understanding user sentiment tendencies are important tasks. This paper is primarily focused on efficiently classifying user comments based on their sentiment tendencies. This study used comments from the health section of Zhihu as the experimental dataset and annotated it for the labeling task. A model combining Bidirectional Encoder Representations from Transformers (BERT) and Long Short-Term Memory (LSTM) was selected for training, and the sentiment classification task was completed. The study discovered that various hyperparameter values and optimizers on the experimental dataset influenced the model’s performance. The accuracy of text sentiment classification for the experimental data set reached 73%. By analyzing the experimental results, the model can effectively complete the task of classifying text in the comment area based on the sentimental state and obtain better classification performance after appropriate optimization steps.

Copyright

© 2024 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - He Li
PY  - 2024
DA  - 2024/10/16
TI  - Addressing Sentiment Classification in Short Text Comments Using BERT and LSTM
BT  - Proceedings of the 2024 2nd International Conference on Image, Algorithms and Artificial Intelligence (ICIAAI 2024)
PB  - Atlantis Press
SP  - 830
EP  - 841
SN  - 2352-538X
UR  - https://doi.org/10.2991/978-94-6463-540-9_83
DO  - 10.2991/978-94-6463-540-9_83
ID  - Li2024
ER  -