Correlating Twitter Sentiment to Bitcoin Market Behaviour

This project intends to find correlations between average sentiment per time interval and BTC market behaviour per corresponding interval

Tech Stack

• REST API
• PostgreSQL
• VADER (Valence Aware Dictionary sEntiment Reasoner)
• NLTK
• Scikit-Learn
• Plotly
• Flask
• Dash

Process:

1. Data Management:
   1. Extract tweets from Twitter target account per given interval => store it in raw tweets db
   2. Preprocess extracted tweets => store it in preprocessed tweets db
   3. Extract BTC stats hourly, daily => store it in db
2. Dashboard:
   1. Hourly/Daily analysis
   2. Display automatically updated graphs

Future Intent

1. Gather sufficient amount of data
2. Perform time series analysis using gathered data