摘要
网络舆情预测是一个非常复杂的非线性问题,传统的反向传播(Back Propagation,BP)神经网络模型在预测过程中存在易陷入局部最优值及收敛速度慢的困境。为了提高预测准确性,笔者结合粒子群算法优化BP神经网络。通过网络爬虫技术获取热点话题的百度指数,并将其作为舆情趋势预测的时间序列数据,对热度值进行预测。通过比较分析,改进后的模型比单一BP神经网络模型的预测精度更高。
Network public opinion prediction is a very complex nonlinear problem.The traditional BP neural network model is prone to fall into local optimum and slow convergence in the process of prediction.In order to improve the prediction accuracy,the author combined the particle swarm algorithm to optimize the BP neural network.The Baidu index of hot topics is obtained through the web crawler technology,and it is used as the time series data of public opinion trend prediction to predict the hotness value.Through comparative analysis,the improved model has higher prediction accuracy than the single BP neural network model.
作者
赵文杰
ZHAO Wenjie(Sichuan Water Conservancy Vocational and Technical College,Chengdu Sichuan 611231,China)
出处
《信息与电脑》
2022年第3期200-203,共4页
Information & Computer
关键词
舆情预测
BP神经网络
粒子群算法
public opinion prediction
BP neural network
particle swarm optimization
