Home
wallstweet is a social media monitoring program that attempts to discover and predict market trends as they relate social media posts. The program was initially developed as a submission for the 24 Hour Edmonton Student Hackathon.
wallstweet mines social media sources (just Twitter for now) and generates a sentiment analysis to try and predict market trends for a specified company. We use machine learning techniques with labelled textual data sources to generate a positive/negative tweet classifier. This classifier is then fed real-time tweet data from a specific slice of time, to try and predict the ensuing asking price for that company's stock.