GitHub - linghton/twitter_trends: Master project for predicting and detecting trends before they happen

Branches Tags

Name		Name	Last commit message	Last commit date
Latest commit History 15 Commits
.DS_Store		.DS_Store
README		README
collection.py		collection.py
collection_trends.py		collection_trends.py
collection_users.py		collection_users.py
dump_uids.py		dump_uids.py
extract.py		extract.py
graphDB.py		graphDB.py
json.py.ck		json.py.ck
output.txt		output.txt
socks.py		socks.py
stats.py		stats.py
test.py		test.py
test1.py		test1.py
test1.pyc		test1.pyc
testing.py		testing.py
testing.pyc		testing.pyc
trends.txt		trends.txt
twitter.py		twitter.py
twitter.pyc		twitter.pyc
twitter_app.txt		twitter_app.txt

Repository files navigation

"
Author :Pranay Agarwal
email  :agarwalpranaya@gmail.com
"
-----------------------------------------------------------------------------------
REQUIREMENTS
    simplejson or json library
    oauth2 : http://github.com/simplegeo/python-oauth2
             Can be installed from above package setup.py script.
twitter_app.txt : contains credentials for the Twitter Applications used.
socks.py        : requred script just include in the same dir.


-----------------------------------------------------------------------------------
SCRIPT COLLECTING DATA

extract.py  : Connect to the timesense, fetch tweets and extract/calculate features
graphDB.py  : Interface to connect to neo4j to store and access social graph

twitter.py  : Interface to make query via TwiterAPI to get user details and followers info

collections.py : Given tweets dump extract author details and gets its followers.
collections_trends.py  : collects trending topics every 5 mins and the relevant popular tweets as well.
collections_users.py  : given follower_uids(unique) dump gets profiles for all of them.