'
Stay hungry Stay Foolish
Analyzed customers’ manner of self-expression on Weibo (Chinese Twitter) content (16GB) and created customers’ persona using Python.
Predicted education models and calculated college distribution in users and friends in MongoDB
Used Genism and Word2Vec to analyze words similarity and build topic models
Used NLTK to build Bigram collection in order to utilize classification corpus to give user’s label
If you are interested in me, please email me to bozhaoli@usc.edu