README

Relations between Twitter and Stocks
By Suvamsh Shivaprasad
ssuvamsh@gmail.com

Description:
- Collected large twitter datasets from live feed and applied noise reduction algorithms
- Using sentiment analysis predicted stock fluctuations for four major technology companies
- Visualized sentiment growth versus real time stock growth

Usage:
This project must be run on the Longhorn Supercomputer at Texas Advanced Computing Center using the visualizaiton portal for easiest compile and run.

1. Load the following modules
	
	module load qt/4.7.0 python/2.7.1-epd paraview

2. Run with the following command:

	vglrun python main.py

3. Google, Apple and Microsoft Stock data: Will show an animated graph of their stock since IPO.

4. Sentiment Analysis: Will display a graph of the growth of sentiment over time per company based on tweet analysis.

5. Compute Word Cloud: Can be used to compute the word cloud for the given twitter dataset.
/****** WARNING THIS WILL TAKE >5 HOURS ON LONGHORN *******/
Hence I have provided precomputed word cloud images

6. Use Display Word Cloud to view the precomputed word clouds for all three companies.