In the selection process of aviation services,users mostly make judgments by viewing the massive amount of comment information.However,with the dramatic increase in comment data and the improvement of data collection methods,the diffi-culty of extracting effective user sentiment information from comments is also increasing.In this paper,a sentiment analysis method based on the BERT model is designed for aviation comment data.On the real aviation comment dataset,the accuracy of this method reaches 0.8141,which can reduce the risk of ticket purchase to a certain extent and improve the service quality of airlines.
关键词
航空评论数据/情感分析/BERT模型
Key words
aviation comment data/sentiment analysis/BERT model
维普
万方数据