针对当前网络伪舆情难以快速、有效辨识的问题,通过对网络舆情特征的分析,认为不同舆情信息反映不同舆情特征,因此辨识的本质可以看作是一个自动分类问题。构建网络伪舆情指标评价体系,提出一种基于支持向量机与K近邻结合的网络伪舆情辨识方法。实验结果表明,该方法比单纯使用支持向量机具有更高的分类精度,且较好地解决了核函数参数选择的问题。 According to the analysis of the characteristics of public opinion on the Internet, the author thinks that different public opinions reflect different characteristics of public opinion. Therefore, the nature of identification can be regarded as an automatic classification problem. Constructing the network pseudo-public opinion index evaluation system, a network-based pseudo public opinion recognition method based on support vector machine and K-nearest neighbor is proposed. The experimental results show that this method has higher classification accuracy than using SVM, and solves the problem of parameter selection of kernel function well.