Skip to content

Latest commit

 

History

History
14 lines (9 loc) · 590 Bytes

readme.md

File metadata and controls

14 lines (9 loc) · 590 Bytes

About

A deep learning-based sentiment analyzer to infer sentiment from the text of short-form financial messages.

Usage

python3 analyze_stock.py stock_ticker

Roadmap

  1. Add more granularity to the time periods that Reddit Fetcher and StockTwits Fetcher can fetch from.
  2. Archive hourly sentiment ratios for each stock using training corpus data.
  3. Graph sentiment scores RELATIVELY! As you would for stock prices! The current absolute system is uninformative because posters tends towards bullish.
  4. Establish correlation score between Sentiment Scores and Stock Prices.