Tibetan microblogging user data analysis and topic identification

The digital era accelerates the construction of information technology, and the Internet penetration rate and the number of Internet users in Tibetan areas have surged, and the value of their online text data is highlighted in cultural heritage, social insights, and technological applications. In this paper, Python is used to capture Tibetan microblog user ID, gender, microblog content, microblog comments and other data, and store them in Hadoop distributed file system to ensure the efficiency and security of data processing. Secondly, the distribution of the number of microblogs, the frequency of microblog posting, the length of microblog content, the proportion of users' gender and the distribution area were analysed, and the results were visualised by using Matplotlib. Then, we compare the effects of LDA and BTM topic models, choose the more effective LDA topic model to further accurately identify and discover the topic feature words in Tibetan microblogs, output the results of Tibetan microblog topic discovery, and analyse the experimental results, which not only enriches the research content in the field of Tibetan information processing, but also provides powerful technical support for the government and institutions in monitoring and analysing the public opinion, guaranteeing the safety of the social network, and responding to the concerns of the public quickly. It not only enriches the research content in the field of Tibetan information processing, but also provides powerful technical support and decision-making basis for governments and institutions in monitoring and analysing social opinions, safeguarding social network security, and responding rapidly to people's concerns.