Mimicry in online conversations

Presenting the paper: do computational measures of linguistic mimicry actually reflect the Interactive Alignment Model? Mostly, but they fall well short of human classifiers on the same data.