SVM和RNN在网络评论情感分析中的比较研究

Comparative Study of SVM and RNN in Emotional Analysis of Online Comments

随着电子商务的迅猛发展,网络评论情感分析研究日益受到重视.分别从传统的机器学习模型和深度学习模型视角,运用支持向量机(Support Vector Machine,SVM)和循环神经网络(Recurrent Neural Network,RNN)方法对向量化表示后的网络评论文本进行情感倾向的学习分析.研究表明,在精确率、召回率及F1等评价指标方面,基于RNN模型的评论情感分析效果明显优于SVM模型.该结果可以帮助消费者更好进行网络消费决策.

With the rapid development of e-commerce,researches in online commentary sentiment analysis have received increasing attentions.From the views of traditional machine learning model and deep learning model,support vector machine(SVM) and recurrent neural network(RNN) were used to analyze the emotional tendency of the network comment text after vectorization.The experimental result indicates that the effect of evaluation sentiment analysis based on RNN model is better than that of SVM model in terms of precision,recall rate and F1 evaluation indicators.The study can help consumers make better online consumption decisions.