摘要
为了保证网络中的海量舆情信息得到高效及时地处理,提出了一种基于数据量的分布式并行SVM混合分类模型。首先通过SVM模型完成舆情文本的类别划分,再根据分类效果反向优化SVM模型,采用朴素贝叶斯模型实现分类效果的提升。构建了一种以数据量为标记的混合反馈式分布SVM模型,根据数据量对所需的子节点数进行动态预测。实验结果表明,提出的模型能够精准分析舆情动向,为舆情监控提供具有重要价值的信息。
In order to ensure the efficient and timely processing of massive public opinion information in the network,a distributed parallel SVM hybrid classification model based on data volume is proposed.Firstly,the classification of public opinion text is completed by SVM model,then the SVM model is optimized according to the classification effect,and the naive Bayes model is used to improve the classification effect.A hybrid feedback distributed SVM model marked by the amount of data is constructed to dynamically predict the number of sub nodes according to the amount of data.The experiment results show that the proposed model can accurately analyze the trend of public opinion and provide valuable information for public opinion monitoring.
作者
王娜娜
WANG Na-na(Department of Network Security,Shanxi Police College,Taiyuan 030401,China)
出处
《信息技术》
2022年第11期14-18,25,共6页
Information Technology
基金
山西省科技厅科学研究基金项目(2020L0716)。
关键词
SVM模型
舆情分析
分布式
模型优化
数据量
SVM model
public opinion analysis
distributed
model optimization
data volume
