基于残差修正的多因素灰色模型的网络舆情预测研究

发布时间：2019-02-26 21:49

【摘要】：【目的/意义】精准预测与掌握舆情事件的发展,及时发现舆情中的潜在危机,对社会的长治久安具有重要意义。【方法/过程】针对网络舆情演化的不确定性、多变性与灰色性等特征,选取多个指标数据建立多因素灰色模型(MGM(1,m))。同时,为提高预测结果的精确度,利用BP神经网络对多因素灰色模型的预测残差进行修正,构建基于残差修正的多因素灰色模型,并结合"莆田系事件"对模型预测性能进行验证。【结果/结论】仿真结果表明,相对于单一序列GM(1,1)模型和无残差修正的多因素灰色模型,残差修正后的多因素灰色模型在网络舆情预测上具有一定的优势。
[Abstract]:Objective: accurate prediction and mastery of the development of public opinion events and timely discovery of potential crises in public opinion are of great significance to the long-term stability of society. [methods / processes] in view of the uncertainty of the evolution of public opinion on the Internet, The multi-factor grey model (MGM (1, m). Was established by selecting multiple index data with the characteristics of variability and grayness. At the same time, in order to improve the accuracy of prediction results, BP neural network is used to modify the residual prediction of multi-factor grey model, and a multi-factor grey model based on residual correction is constructed. Combined with Putian event, the prediction performance of the model is verified. [results / conclusion] the simulation results show that compared with the single sequence GM (1,1) model and the multi-factor grey model without residual correction, the prediction performance of the model is verified. The multi-factor grey model with residual error correction has some advantages in the prediction of network public opinion.
【作者单位】：福州大学经济与管理学院;
【分类号】：C912.63

【相似文献】