Sentiment analysis of Chinese comments on OTA website using  BERT and LSTM

Juan Shen; Mei Xu

doi:10.1117/12.2642660

28 July 2022 Sentiment analysis of Chinese comments on OTA website using  BERT and LSTM

Juan Shen, Mei Xu

Proceedings Volume 12303, International Conference on Cloud Computing, Internet of Things, and Computer Applications (CICA 2022); 123031K (2022) https://doi.org/10.1117/12.2642660
Event: International Conference on Cloud Computing, Internet of Things, and Computer Applications, 2022, Luoyang, China

Abstract

Under the background of "Internet + tourism", many OTA sites have large numbers of users and have accumulated huge amounts of user review information. In this paper, web crawler is used to grab Chinese comment data on OTA website for pre-processing. BERT pre-training model is used to represent text feature vectors at sentence level. The results are then input into LSTM model to obtain high-level semantic features. Thus, the emotional tendency of the comments can be classified. The comparison of experimental data shows that BERT-LSTM model has higher classification accuracy, recall rate and F1 value. It shows that the effectiveness of text sentiment analysis has been improved. An emotional analysis of these emotional commentary texts is conducive to the monitoring of management departments, providing reference for the government and enterprises to make decisions. And this research has very high application value for promoting the development of tourism.

Citation Download Citation

Juan Shen and Mei Xu "Sentiment analysis of Chinese comments on OTA website using  BERT and LSTM", Proc. SPIE 12303, International Conference on Cloud Computing, Internet of Things, and Computer Applications (CICA 2022), 123031K (28 July 2022); https://doi.org/10.1117/12.2642660

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available