Twitter data mining - social media as a rising customer service channel

• Presentation slides ---- Thesis paper

• Collect data from Twitter API using Python scripts that records the customer service interactions on major consumer electronic brands such as Apple and Samsung

• Parse Json data objects in Python and organize tweets into customer service sessions with unique customers with Pandas

• Use NLTK(Natural Language Processing toolkit) to conduct Text Mining and Sentiment Analysis

• Use Naive Bayes classifier and bag of words method to predict sentiment of a customer towards the customer service received

Unfortunately my codes are forever gone with damaged computer, and I was too inexperienced to back it up on github.