Currently in creating the mentionball we request both tweets and RT's from each user in the ball (include_rts=1 in the request). But then we throw them away in the process of creating the final graph.
If we want to keep collecting RTs, then we should use them somehow. One option is to give each person in the ball a "reach" based on the number of followers of those that retweet them. This is a very crude measure used by some Twitter analytics browser plugins. There's a lot wrong with it as far as an estimation of influence goes but it could be better than nothing.
Another option is to include RT's as an alternative kind of edge between nodes, labeled appropriately.
Looks like these RT's count against the maximum tweets returned whether included in the response or not so no need to collect them.
Currently in creating the mentionball we request both tweets and RT's from each user in the ball (include_rts=1 in the request). But then we throw them away in the process of creating the final graph.
If we want to keep collecting RTs, then we should use them somehow. One option is to give each person in the ball a "reach" based on the number of followers of those that retweet them. This is a very crude measure used by some Twitter analytics browser plugins. There's a lot wrong with it as far as an estimation of influence goes but it could be better than nothing.
Another option is to include RT's as an alternative kind of edge between nodes, labeled appropriately.
Looks like these RT's count against the maximum tweets returned whether included in the response or not so no need to collect them.
https://dev.twitter.com/docs/api/1.1/get/statuses/user_timeline