Home

OK, it may be complicated, as a lot of stuff is still up in the air, but this is the very basic help to have the system initialized and running. What is all about? the short version: geo-clustering and topic extraction from twitter. Theoretically the same algorithms can be used on different data sources: the database and code are designed to allow some flexibility, given enough resources to refactor a bit of code and add some functionality.

Please refer to the following pages on this wiki to have your system up and running:

• Installation: how to have the system ready to go
• Backend: documentation for fetching and clustering data with the java algorithms implemented in this distribution (it's where part of the magic happens)
• Frontend: data-viz functions, where the other part of the magic happens
• API: a list of RESTful services to build your own front-end
• Cron jobs: useful maintenance scripts to keep the system live and up to date

This is a recent screenshot of the topic stats functionality.

There are 2 core elements in this deployment:

• archive fom.jar is a collection of algorithms and tools for fetching and analyzing data
• subfolder /php under the main git root contains a CodeIgniter based application for data viz