基于大数据驱动的媒体风险感知模型

Media Risk Perception Model Driven by Big Data

大数据背景下媒体历史数据隐藏丰富信息,通过分析和挖掘这些信息,可以掌握媒体风险的变化规律,传统模型无法对媒体风险的变化规律进行准确刻画,导致媒体风险感知误差大,为了获得理想的媒体风险感知效果,设计了基于大数据驱动的媒体风险感知模型.首先采集媒体历史数据,并对媒体历史数据进行预处理,然后引入大数据驱动的最小二乘支持向量机对预处理后的媒体历史数据进行建模,建立媒体风险感知模型,并通过粒子群算法对媒体风险感知模型参数进行优化,最后以某媒体风险数据为例,与传统媒体风险感知模型进行对比,结果表明,本模型的媒体风险感知精度超过95%,不仅远远高于传统模型的媒体风险感知精度,而且具有较强的通用性.

Under the background of big data, media historical data hides rich information. By analyzing and mining this information, we can grasp the change law of media risk. The traditional model cannot accurately describe the change law of media risk, resulting in large error in media risk perception. In order to obtain the desired effect of media risk perception, a media risk perception model driven by big data is designed. Firstly, the media historical data is collected and preprocessed, then the least squares support vector machine driven by big data is introduced to model the preprocessed media historical data, to establish the media risk perception model, and to optimize the parameters of the media risk perception model through particle swarm optimization algorithm. Finally, a media risk data is taken as an example to be compared with the traditional media risk perception model. The results show that the media risk perception accuracy of this model is more than 95%, which is not only much higher than that of the traditional model, but also more versatile.