Interesting that most hashtags don't get more than 3 posts per hour - that breakdown would be a good graph to see! It's also curious that they don't apply some type of seasonal decomposition to the timeseries data; though perhaps that that makes sense with Instagram's particular data.
Comments
This was quite good.
Interesting that most hashtags don't get more than 3 posts per hour - that breakdown would be a good graph to see! It's also curious that they don't apply some type of seasonal decomposition to the timeseries data; though perhaps that that makes sense with Instagram's particular data.
Google [zipfian], and I bet that's what the distribution looks like...