论文部分内容阅读
针对网络伪舆情的识别问题,提出一种基于支持向量机的网络伪舆情识别方法。鉴于不同的舆情信息所反映出的舆情特征不同,而舆情特征的不同又可进一步辨别舆情的真假,因此首先构建针对网络舆情真伪的评价指标;基于支持向量机的分类机理,结合网络舆情的评价指标提出基于支持向量机的网络伪舆情识别模型,采用多项式核函数以及优化之后的径向基核函数产生的分类器。通过实验证明采用支持向量机构造舆情分类器所构建的识别算法能够对网络伪舆情进行有效识别。
Aiming at the problem of network pseudo-public opinion identification, this paper proposes a method of network pseudo-public opinion recognition based on support vector machine. In view of the different characteristics of public opinion reflected by different public opinion information, and the different characteristics of public opinion can further distinguish the true and false of public opinion, we first build the evaluation index for the authenticity of network public opinion. Based on the classification mechanism of support vector machine, This paper proposes a network pseudo-public opinion recognition model based on SVM, a classifier generated by polynomial kernel function and radial basis function after optimization. Experiments show that using SVM to construct the recognition algorithm based on public opinion classifier can effectively identify the network pseudo public opinion.