基于BP神经网络和遗传算法的网络舆情危机预警研究  

On Network Public Opinion Crisis Early Warning Based on the BP Neural Network and Genetic Algorithm

作　　者：孙玲芳[1] 周加波 林伟健[1] 候志鲁 许锋[1] 

机构地区：[1]江苏科技大学经济管理学院,镇江212003

出　　处：《情报杂志》2014年第11期18-24,共7页Journal of Intelligence

基　　金：教育部人文社会科学基金资助项目"基于大众标注集体智慧的网络群体事件主题分类研究"(编号:10YJAZH069);江苏省"六大人才高峰"高层次人才项目"网络化集体智慧及其在群体事件预警中的应用研究"(编号:XXRJ-013)

摘　　要：Web2.0时代,如何对网络舆情危机进行有效预警已经成为政府部门的必修课。本文充分考虑了网络舆情危机产生、发展、变化的规律及特点,综合现有指标体系的优缺点,建立了3个一级指标和11个二级指标的网络舆情危机预警的指标体系。利用遗传算法优化BP神经网络的初始权值与阀值,构建了基于BP神经网络和遗传算法的网络舆情危机预警模型。最后,通过仿真实验,结合5个具体案例对该模型进行了验证与分析。实验表明,本文建立的网络舆情预警指标体系与遗传BP神经网络模型是有效可行的,预警准确率要优于标准的BP神经网络网络模型。In a Web2. 0 era, how to conduct an effective early-warning of the network public opinion crisis has become a required course for government departments. This paper sufficiently considers the development, changes in laws and characteristics of the network public opinion crisis and establishes a 3-level early warning index system with 11 secondary indexes of network public opinion crisis. Then,by u-sing genetic algorithm to optimize the BP neural network＇s initial weights and thresholds, anetwork public opinion crisis early warning modelis constructed. Finally, the model is verified and analyzed through a simulation experiment involving5 specific cases. Experiment shows theearly warning index system of network opinion and the genetic BP neural network model establishedare effective and feasible, and the warning accuracy is superior to the standard BP neural network model.

关 键 词：BP神经网络 遗传算法 网络舆情 舆情预警 预警指标 

分 类 号：C930[经济管理—管理学]