Research on Weibo user behavior system for subjective perception and big data mining technology

Abstract

Firstly, the paper analyzes the information propagation mode of Sina Weibo, and proposes a new MURank algorithm based on PageRank algorithm, which fully considers the relationship between users, the transmission relationship between users and microblogs, and the forwarding relationship between microblogs. At the same time build the user-microblog mode diagram. The design of the thesis is also more in line with the information dissemination mode of the actual Weibo platform. Experiments prove that MURank can effectively solve the impact of “zombie powder” and “level sinking” problems, and more realistically reflect the level of influence of users. Secondly, the paper combines BP neural network algorithm to improve the QoE hierarchical index system, and proposes a five-level index model. The mobile video service is taken as an example to illustrate the establishment process of the indicator system and the QoE quantitative evaluation method. By verifying on the open source distributed platform Hadoop platform, it is found that the proposed method is a solution to deal with massive data analysis.