Real time sentiment analysis on Twitch chat messages.
This project gives shape to a tool called Twitch Chat Listener, capable of performing real-time sentiment analysis and providing streamers with the ability to analyze, extract knowledge, react to the polarity of sentiment (rejection, indifference or welcome) of the comments that their viewers are sharing through the streaming chat and, ultimately, maximize audience loyalty and revenue generation.
Keywords: Twitch; live streaming; streamer; IRC chat; natural language processing (NLP); sentiment analysis; non-relational database; web scrapping; dashboard; Docker containers.