首页> 外文期刊>Journal of Residuals Science & Technology >Network Public Opinion Analysis Method Based on SVM Non Kernel Parameter
【24h】

Network Public Opinion Analysis Method Based on SVM Non Kernel Parameter

机译:基于支持向量机非内核参数的网络舆情分析方法

获取原文
           

摘要

Network public opinion prediction is a kind of complex prediction issue of poor information, small sample and uncertainty and a kind of network public opinion prediction model based on grey support vector machine is established in order to improve the accuracy of network public opinion prediction. Firstly, text clustering, hotspot extraction, data aggregate and other preprocessing are carried out for network data, then GM (1,1) model of time sequence of network public opinion is established and support vector machine is used to correct the predicted result of GM (1,1) model, and finally, model performance is tested via simulation experiment. Simulation result shows that, relative to traditional prediction model, grey support vector machine model improves the prediction accuracy of network public opinion and the predicted result is of certain practical value.
机译:网络舆情预测是信息量少,样本少,不确定性复杂的预测问题,为了提高网络舆情预测的准确性,建立了一种基于灰色支持向量机的网络舆情预测模型。首先对网络数据进行文本聚类,热点提取,数据聚合等预处理,然后建立网络舆情时间序列的GM(1,1)模型,并利用支持向量机对GM的预测结果进行校正。 (1,1)模型,最后,通过仿真实验测试了模型性能。仿真结果表明,相对于传统的预测模型,灰色支持向量机模型提高了网络舆情的预测精度,预测结果具有一定的实用价值。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号