Paper
16 December 2022 Research on rumor detection based on BERT
Author Affiliations +
Proceedings Volume 12500, Fifth International Conference on Mechatronics and Computer Technology Engineering (MCTE 2022); 1250043 (2022) https://doi.org/10.1117/12.2660814
Event: 5th International Conference on Mechatronics and Computer Technology Engineering (MCTE 2022), 2022, Chongqing, China
Abstract
With the fast-speeding development of the information age, social media also provides a central channel for people to obtain information. However, the mixed-up information provided by various network platforms is hard to identify, the negative aspects of public sentiment and the spread of the rumor also make significant aspects in cultural ecological environment. At present, rumor detection is mainly focusing on deep learning, extracts and analyzes the text semantic features and then makes predictions. But, using this method to select eigenvalues always lacks varieties, it ignores the semantic integrity and the potential relationship between data. In order to improve the efficiency and accuracy in rumor detection, this research provides a method in predicting and analyzing text features, which based on the pre-training model of BERT, using the public Weibo-20 and Weibo-21 datasets, tunes the model parameters through the comparative experiments, the result shows that the processing speed and effectiveness is far more fruitful.
© (2022) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Yi Zhao and Mingshu Zhang "Research on rumor detection based on BERT", Proc. SPIE 12500, Fifth International Conference on Mechatronics and Computer Technology Engineering (MCTE 2022), 1250043 (16 December 2022); https://doi.org/10.1117/12.2660814
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Transformers

Computer programming

Data modeling

Internet

Web 2.0 technologies

Analytical research

Feature extraction

RELATED CONTENT


Back to Top