regex character groups, duh. not at the part where weighting of metrics is being considered, but in general i feel like this is relatively low weight metric. even when you detect punctuation, what should the metric be. mean or median number of punctuation characters included? percentage of messages with punctuation? Is it all punctuation, or just commas?
regex character groups, duh. not at the part where weighting of metrics is being considered, but in general i feel like this is relatively low weight metric. even when you detect punctuation, what should the metric be. mean or median number of punctuation characters included? percentage of messages with punctuation? Is it all punctuation, or just commas?