【摘要】 微博的传播热度研究对加强舆情监控、提高市场营销效率等具有重要作用。设计基于LDA算法提取微博主题特征,并融合热点话题等其他特征挖掘用户关系网络中的"隐形粉丝"信息,将传播深度和传播广度特征作为衡量微博传播效果的重要指标,最后结合BP神经网络建立微博热度预测模型。实验结果表明,加入间接用户关系网络信息和主题信息能够有效地提高微博热度预测模型的性能,在准确率、召回率等指标值上都有较好的提高,验证了算法的有效性。更多还原

【Abstract】 The research on the popularity of microblog message plays an important role in strengthening the monitoring of public opinion and improving the efficiency of marketing. In view of this,an algorithm is proposed in this paper. In the algorithm,the topic features of the microblog are extracted on the basis of LDA(latent Dirichlet allocation)algorithm,and then the hot topics and other features are integrated. The information of ″ invisible fans″ in the user relationship network is mined.The features of propagation depth and propagation width are taken as the important indicators to measure the propagation effect of microblog. The microblog popularity prediction model is constructed in combination with the BP neural network. The experimental results show that,in comparison with the other algorithms,the proposed algorithm can effectively improve the performance,accuracy,recall rate and other index values of the popularity prediction model by adding indirect user relationship network information and topic information,which has verified the effectiveness of the algorithm.更多还原