I use tagging to simplify a wide range of operations in my application. One of the things I do is tag a number of topics that appear in a tweet and increment how many times these topics have been mentioned in the same tweet as links that we have gathered through the links augmentation.
For example, if someone shared a post for the first time by amazon.com about it’s new tuition assistance program along with certain keywords indicating they were talking about education and jobs, the link would be added to the database along with an array of tags and how many times this same url had been mentioned in concert with each tag. If another person mentioned the same exact url using only keywords that related to education, only the education tag would be incremented. This has been tested pretty extensively and gives us an excellent way to separate out chaff when someone is carpet bombing a link by mentioning it in reference to topics we track (the twitter handles of multiple only vaguely related politicians in most cases). We also store the links.retweet_count element of each link and update this on each additional inclusion of the link.
My problem is that the counts for the tags are almost always higher than the retweet_count of the link when most recently seen. This leads me to believe that the description of links.retweet_count, “The total number of times this link has been Tweeted” is incorrect, at least for the data delivered (I can’t say for sure that it is incorrect as far as querying). Can you tell me what this number actually denotes?