网络热门话题的跟踪建模与检测分析  

Tracking modeling and detection analysis of network hot topics

作　　者：李晶[1] 

机构地区：[1]广西财经学院,广西南宁530003

出　　处：《现代电子技术》2018年第3期128-131,共4页Modern Electronics Technique

摘　　要：网络热门话题的跟踪对社会稳定、国家安全具有重要作用,针对传统网络热门话题跟踪模型误差大的缺陷,提出改进神经网络的网络热门话题的跟踪和检测模型。首先对网络热门话题的跟踪和检测现状进行分析;然后收集网络热门话题相关数据,采用神经网络对网络热门话题的跟踪进行建模,实现网络热门话题的检测,并对神经网络的局限性进行改进;最后采用具体网络热门话题的跟踪和检测实验对模型的有效性进行测试,与其他网络热门话题的跟踪和检测模型相比,改进神经网络提高了网络热门话题的跟踪精度,降低了网络热门话题的检测错误率,可以为网络舆情管理者提供可靠的信息。The tracking of network hot topic plays an important role in social stability and national security.Since the tracking model of the traditional network hot topics has large error,a tracking and detection model of the network hot topics based on improved neural network is proposed.The tracking and detection status of the network hot topics is analyzed,and then the relevant data of the network hot topics is collected.The neural network is used to model the network hot topics tracking,and detect the network hot topics.The limitation of the neural network is improved.The effectiveness of the model is tested with the tracking and detection experiments of specific network hot topics.In comparison with other tracking and detection models of network hot topics,the improved neural network model can improve the tracking accuracy of the network hot topics,reduce the error detection rate of the network hot topics,and provide the reliable information for the network public opinion managers.

关 键 词：网络热门话题 话题跟踪 话题检测 跟踪与检测模型 神经网络 检测错误率 

分 类 号：TN711-34[电子电信—电路与系统]