Author : Karthik Anantha Padmanabhan
This will be a set of tools that can be used to analyze live football(soccer) matches using twitter as the medium. The tasks include event summarization , automated player ratings and some more that will added.
- Version 1.6 of the Java 2 SDK (http://java.sun.com)
TF-pundit uses SBT (Simple Build Tool) with a standard directory structure.
./build update compile
##Summarization and Player Performance Analyzer
These are the two components of the system that have currently been implemented. To use this, first identify the keywords
associated with a particular game. For example, for a game between AC Milan and Catania the keywords maybe acm, acmilan,catania,acmcatania,milancatania
To run TF-Pundit :
$ bin/fanalyst run footballTwitter.twitter.GameAnalyst -t "game keywords" -p "players to follow"
Replace players to follow with the player whose performance is to be monitored
Example :
bin/fanalyst run footballTwitter.twitter.GameAnalyst -t "milan" "barca" "acm" "barca" -p "messi" "balotelli" "xavi"
The above command can be used to follow a game betweebn AC Milan and FC Barcelona. The players who will be analyzed are
messi balotelli and xavi
The current implementation produces text files having the summary in rtSummary.txt
and the sentiment associated with the players in scores.txt.
The tweets location(geo-coordinates) are also written to location.txt