基于组合预测的网络舆情数据预测模型研究  

Research on the Prediction Model of Network Public Opinion Based on Combination Forecasting Method

作　　者：徐敏捷[1] 兰月新[1] 刘冰月[2] 

机构地区：[1]中国人民武装警察部队学院,河北廊坊065000 [2]河北工业大学城市学院,天津300132

出　　处：《情报科学》2016年第12期40-45,87,共7页Information Science

基　　金：国家社科基金青年项目(15CXW015);全国统计科学研究计划重点项目(2014LZ10);国家社会科学基金重大项目(14ZDA063)

摘　　要：针对网络舆情传播过程中引导与监督不及时的现状,通过组合预测模型,将常用的数据预测模型Logistic模型、指数平滑法模型和灰色GM(1,1)模型结合在一起,利用AHP法进行权重赋值,得出最优化网络舆情数据的预测值。并通过案例进行比较分析,对于网络舆情管控,维护社会稳定具有重要作用。At present, guidance and supervision is not a timely dissemination process for the network of public opinion. The paper combines forecasting model to predict network public opinion data. It combines the common data forecast model of lo- gistic model, exponential smoothing model and grey GM （1,1） together, and the weight assignment is based on AHP method to get the optimal network public opinion data value for applying to the analysis and comparison of the actual situation. The research of this paper will be in favor of controlling public opinion and maintaining social stability.

关 键 词：组合预测 网络舆情 LOGISTIC模型 

分 类 号：G206.3[文化科学—传播学]