Sentiment Analysis for Crypto Meme Tokens

A comprehensive sentiment analysis system for analyzing social media sentiment and its correlation with cryptocurrency token prices, specifically focused on meme tokens.

Project Structure

├── database/               # Database management scripts
├── models/                 # ML models including GAN implementation
├── ohlcv_data/            # Raw OHLCV data for various tokens
├── ohlcv_data_standardized/# Standardized OHLCV data
├── scripts/               # Core analysis scripts
└── twitter/               # Twitter data collection scripts

Key Features

• Custom VADER sentiment analysis tuned for crypto and meme-specific terms
• Real-time token price data collection via DexRabbit
• Twitter data scraping and analysis
• Time series sentiment aggregation
• SQLite database for efficient data storage
• GAN model for synthetic data generation

Setup

1. Clone the repository
2. Install dependencies:

   pip install -r requirements.txt

3. Create a .env file with your API credentials:

   DEXRABBIT_EMAIL=your_email
   DEXRABBIT_PASSWORD=your_password
   

Usage

1. Data Collection:

   python twitter/data_collector.py  # Collect Twitter data
   python twitter/fetch_tokens.py    # Fetch token prices

2. Sentiment Analysis:

   python scripts/meme_sentiment_analyzer.py

3. Data Standardization:

   python scripts/standardize_ohlcv.py

Database Schema

The project uses a single SQLite database (sentiment_data.db) containing:

• Tweet content and metadata
• Author information
• VADER sentiment scores
• Token mappings
• Sentiment time series data

Contributing

Feel free to submit issues and enhancement requests!