面向负面网络舆情的识别与追踪关键技术研究
Research on key technologies for the identification and tracking of negative network public opinions
李学威 1孙滨2
作者信息
- 1. 周口职业技术学院信息工程学院,周口 466000
- 2. 郑州工业应用技术学院信息工程学院,郑州 451150
- 折叠
摘要
当前网络负面舆情的追踪技术只完成了以关键词为基础的话题区分,对内部数据的相似性没有进一步研究,子话题的划分过程也过于简单.为了实现对网络负面舆情精准地识别和追踪,研究面向负面网络舆情的识别与追踪关键技术.首先,划分出负面网络舆情的子话题,分析数据对象的相似特性、完成聚类,获得一组子话题聚类的集合;其次,识别负面网络舆情事件,即从海量的事件信息中挖掘并识别出事件的本质,推断事件的因果关系,并对其发展趋势进行预测;最后,实现负面网络舆情话题的追踪.实验结果表明,此方法对负面网络舆情的识别检查的误检率平均值为1.1%,漏检率平均值为0.9%,通过实验结果能够得出此方法在负面网络舆情识别与追踪中具有较高的准确性和可靠性.
Abstract
At present,the tracking technology of network negative public opinion only completes the topic distinction based on keywords,without further research on the similarity of internal data,and the division process of sub-topics is too simple.In or-der to realize the accurate identification and tracking of negative public opinions,the key technology of identification and tracking of negative public opinions is studied.First,divide the sub-topics of negative network public opinion,analyze the similar character-istics of data objects,complete the clustering,and obtain a set of sub-topic clusters;second,identify negative network public opin-ion events,that is,dig and identify the essence of the event from the massive event information,infer the causality of the event,and predict the development trend;finally,realize the tracking of negative network public opinion topics.The experimental results show that the average misdetection rate of negative network public opinions is 1.1%and the average detection rate is 0.9%.The ex-perimental results can show that the method has high accuracy and reliability in the identification and tracking of negative network public opinions.
关键词
负面网络舆情/舆情追踪/舆情识别/数据聚类/数据挖掘Key words
negative network public opinion/public opinion tracking/public opinion identification/data clustering/data mining引用本文复制引用
基金项目
河南省科技厅科技攻关支持项目(232102210200)
河南省科技厅科技攻关支持项目(222102210159)
河南省科技厅软科学支持项目(222400410228)
河南省教育厅高等学校重点科研项目(23B520036)
出版年
2024
NETL
NSTL
万方数据