微型电脑应用

在航空服务的选择过程中，用户通常通过查看海量评论信息进行判断。然而，随着评论数据剧增和数据采集手段的提升，如何快速从评论中提取有效用户情感信息的难度也不断上升。基于BERT模型设计出一种针对航空评论数据的情感分析方法。在真实的航空评论数据集上，该方法准确率能够达到0.8141,可在一定程度上降低购票风险，提升航空公司服务质量。

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 difficulty 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.