基于方面级的餐厅用户评论细粒度情感分析

Fine-particle Sentiment Analysis Based on Aspect Level Restaurant Customer Comment

评论数据的情感分析一直是自然语言研究的热点之一,特别是评论观点丰富性、情感化、多元化、非结构化等特征方面的研究近年来深受大家关注。本文基于AI Challenger2018细粒度情感分析比赛为研究背景,在分析GCAE和SynATT两种模型基础上,通过研究方面类别情绪分析(ACSA)方法,提出了CNN-GCAE和CNN-SynATT模型,解决了原来模型在数据处理方面的不足,提高了情感分析的精准度和召回率。实验结果表明,改进模型对评论数据情感分析的准确率效果明显。

The emotion analysis of comment data has always been one of the hot topics in the study of natural language, especially the research on the richness, emotion, diversity and unstructure of comment views. Based on the AI Challenger2018 fine-grained emotion analysis competition as the research background, this paper challenges two models, GCAE and SynATT, and proposes the CNN-GCAE and CNN-SynATT models through the research category emotion analysis(ACSA) method. It solves the shortage of the original model in data processing and improves the accuracy and recall rate of emotion analysis. The experimental results show that the improved model has a significant effect on the accuracy of emotional analysis of critical data.