论文部分内容阅读
从我国网络舆情现状出发,分析网络舆情对社会影响的主要因素,选取4种网络舆情预警指标,获取观测数据,基于改进的D-S证据理论信息融合技术,建立网络舆情预警模型,并利用类概率函数实现对网络舆情话题演变成舆情事件的可能性的判断。同时把改进后的融合方法与经典的D-S组合规则做比较,结果表明,改进后的方法可以更好地实现证据的融合,并得到更为合理的判断结果。
Based on the current situation of Internet public opinion in our country, this paper analyzes the main factors that affect the public opinion on the Internet, selects four types of Internet public opinion early warning indicators, and obtains the observed data. Based on the improved DS evidence theory and information fusion technology, an early warning model of Internet public opinion is established. To achieve the network of public opinion topics into the possibility of public opinion events judgment. At the same time, the improved fusion method is compared with the classical D-S combination rule. The results show that the improved method can better achieve the fusion of evidence and obtain a more reasonable judgment.