Cher Ami: A Tweet Classifier

Cher Ami is a project that analyses and classifies tweets from Twitter. This project is an assignment for CS4242 Social Media Computing, National University of Singapore.

The project is named after a homing pigeon that saved many lives during World War I in 1918.

Installation

Installation is not required. However, the following dependencies are required to run the classifier:

• tweepy, a library for Twitter
• iso8601, a library to parse ISO 8601 formatted dates

Configuration

The training set is provided as a file.

There are sources where tweets can be obtained:

• Link (A status link is supplied)
• Twitter (Uses Twitter's Streaming API)
• File (Uses JSON data from a file)

To use Twitter's Streaming API, one needs to be registered on Twitter Developers in order to obtain the required information:

• OAuth consumer key
• OAuth consumer secret
• OAuth access token key
• OAuth access token secret

Copy cherami/config.py.default to cherami/config.py and modify cherami/config.py accordingly.

Running

The classifier can be run by using the following command:

python cherami/cherami.py

Documentation

There is no documentation provided at the moment.