initial_beta_twitter_information_provenance
python frameprov.py
This is the initial idea of using machine learning libraries to perform analysis on tweets. Using TwitterAPI to pipline tweets from twitter about any topic that we choose and analyzing features of the tweets and meta data. Due to restirctions of use in twitterAPI very minimal usages can be seen on the system since twitter only let's you use 100.
As the initial stage this product has only one feature that is using a certain key word from a topic of intrest it gives how many people out of 100 are talking positive about that topi.
For example the system usages default topic word 'Family guy' and outputs the number of people out of 100 that are talking good things about family guy in twitter. The system is designed to take any topic word as an argument and return analtics over that word but due to a bug API query that system is currently not able to do that only as of now.
*Working Python2.7 or above
*NLTK library for machine learning