weibo-analysis-and-visualization

Why we included this project

This one is a solid working reference for anyone analyzing Chinese-language social media, since it takes Weibo posts and comments and carries them through preprocessing, traditional-to-simplified conversion, and a long list of analyses in one script set. The visualization side covers word clouds, heat maps, time trends, and relationship graphs, while the NLP side includes LDA topic modeling, dictionary and neural (word2vec plus LSTM) sentiment analysis, TF-IDF and word-vector clustering, keyword extraction, and text similarity. It expects MongoDB JSON dumps from the WeiboSpider crawler, so it reads best as a practical walkthrough of the whole pipeline rather than something you import and call. Analysts and students building Chinese NLP pipelines can lift individual scripts as templates, with the caveat that you edit hardcoded file paths by line number before anything runs.