随着信息时代的到来,网络评论数量急剧增加,对于这些网络评论的倾向性分析是网络舆情研究中的重要课题.为了对网络评论的倾向性进行分析,提出了基于神经网络的网络舆情文本倾向性分析算法,使用神经网络训练的方法构建了较好的情感词倾向值词典,对网络评论的倾向性预测达到了较高的准确率.对算法进行了细致的描述,同时提出了基于神经网络的网络舆情文本倾向性分析系统框架.最后通过实验进行测试,证明了该系统能够达到较好的准确率.%With the Information age coming, the amount of network comments increase greatly. Text Orientation Analysis on network reviews is one of the most important projects in public opinion research. In order to analysis Text Orientation, this paper proposes a Text Orientation Analysis Algorithm based on Neural Network; create a good orientation dictionary using Neural Network training. The paper describes the algorithm in detail, and proposes the system framework. Finally, this paper performs an experiment to prove that the system can achieve a better accuracy.
展开▼