Twitter Driven News Generation
Using social-media for news and events is an especially fast and effective way of collating information from many distributed sources. However such information is often unstructured, difficult to parse and even contradictory.
Feeds from Twitter also have interesting and bursty performance characteristics. During large sporting tournaments or news events, many 1000s of messages can be posted per second.
This project has many aspects - but has a central goal of being able to collate large amounts of data and present it in a coherently fashion.
The exact presentation will depend on the event: a vote, a tournament, an election, a televised debate, even a popular uprising. You will have to identify as many different possible modes of operation as possible and write a server to extract and present that information dynamically, link it to other information sources, and produce a result that captures the spirit of the event.
- Use Trello for development board
- Used AlchemyAPI for keyword extraction
Create a config file called config.conf.
Into it, for now put:
[twitter] username = username password = password [alchemyapi] key = apithing