Research on Prediction the Development Trend of Weibo Topics Based on Fuzzy Neural Network
Ruijie Miao, Kongyu Yang · Journal of Physics Conference Series · 2019
Abstract The people who use the Weibo application are widely available, and Weibo has the characteristics of instant openness, strong interaction, fast transmission speed and wide influence range. Therefore, predicting the development trend of Weibo topics quickly and accurately is a topic worth studying. In this paper, the total number of posts in a Weibo topic is used as a quantitative indicator of the development the Weibo topic. Fuzzy neural network is used to predict the hot trend of Weibo topic, and the improved annealing algorithm is used to optimize the membership function of the fuzzy neural network and its quantity to improve the prediction accuracy. The superiority of the proposed method is verified by the comparison experiment of the collected Sina Weibo dataset. The method proposed in this paper effectively solves the problem that the parameters in the membership function are complex and easy to fall into the local optimum, which improves the accuracy of the development trend of Weibo public.