自然语言处理中的情感分析技术  

Emotional Analysis Techniques in Natural Language Processing

作　　者：李勇 王浩 LI Yong;WANG Hao(Chinese People's Liberation Army Aviation School,Beijing 101116,China;Tangshan Normal University,Tangshan,Hebei 063002,China)

机构地区：[1]中国人民解放军陆军航空兵学院,北京101116 [2]唐山师范学院,河北唐山063002

出　　处：《移动信息》2023年第12期165-166,176,共3页MOBILE INFORMATION

摘　　要：文中探讨了在自然语言处理领域中,利用长短期记忆神经网络(LongShort-Term Memory,LSTM)与朴素贝叶斯方法进行情感分析的技术。通过对IMDb影评数据集的实验,深入研究了特征提取与选择、LSTM和朴素贝叶斯的结合在情感分析中的应用。实验结果表明,该方法在分类准确率、精确率、召回率等指标上表现出色,相较于单独使用LSTM的方法,具有更好的性能。这表明,将LSTM的判别性能力与朴素贝叶斯的概率建模相结合,对情感分析任务的提升具有显著影响,为情感分析技术的进一步发展和实际应用提供了有益的参考。In the field of natural language processing,this paper discusses the technology of sentiment analysis using long short-term memory neural networks(LSTM)and naive Bayes method.Through experiments on IMDb film review datasets,the application of feature extraction and selection,LSTM and naive Bayes in sentiment analysis is deeply studied.The experimental results show that the method performs well in classification accuracy,precision,recall and other indicators,and has better performance than the method using LSTM alone.This shows that combining the discriminative ability of LSTM with the probabilistic modeling of naive Bayes has a significant impact on the improvement of the sentiment analysis task,providing a useful reference for the further development and practical application of sentiment analysis technology.

关 键 词：自然语言处理 情感分析 支持向量机 朴素贝叶斯 

分 类 号：TP391.1[自动化与计算机技术—计算机应用技术]