tweet_classification (WIP)

An Attempt to perform simple classification on Tweets that is fetched from a tweet stream.

Uses sklearn for the classification and tweepy for the Twitter api.

Running the code

For running this code, you need to have a twitter dev account. After creating the account you need to clone this repo and make a directory in the same folder named _config.py In this file, please enter the follwowing

_config = {
‘consumer_key’ :’’, # your consumer_key
‘consumer_secret’ : ‘’, # your consumer_secret
‘access_token’ : ‘’, # your access_token
‘access_secret’ : ‘’ # your access_secret
}

That’s it!

To create training data, run the dump_tweets.py with the stream names as arguments. So for example if you need to get streams of ‘basketball’ and ‘football’, you type:

python dump_tweets.py basketball football

Citations

I referred the following sources

• tweepy : http://sachithdhanushka.blogspot.in/2014/02/mining-twitter-data-using-python.html

Built With

• python
• scikit-learn
• shell
• tweepy