Research on User Influence Based on Advanced LAD: Taking Baidu Tieba as an Example

A huge amount of information is spread in the network, and each kind of information contains the possibility of influencing public opinion. However, considering the number of reasons, it is not realistic to review the information transmitted by users. It is of practical value to both the development of enterprises and the work of the government to analyze the network contents that are likely to become popular information in the future by means of prediction. This paper makes an in-depth study on the relevant theories of online community user influence and network content popularity, and summarizes the methods and shortcomings of domestic and foreign mainstream research. On this basis, the user influence measurement model based on PageRank and LDA and the post popularity measurement model based on neural network are constructed. In this paper, the Topic factor is innovatively brought into the measuring range of user influence, and the topic-relation Rank measurement model of user influence is established based on PageRank. Aiming at Baidu Tieba, the article uses crawlers to obtain real data and conduct experiments. The results show that the user influence measurement model integrates three influencing factors, namely user, network topology and text topic, so the model has a good measuring effect.