It's interesting to me how the authors' CAC approach was able to find a whole new (and significant) cluster, partisan taunting, in congressmen's press releases, but was not able to detect an important cluster, 9/11, in presidential addresses. Is it because the latter is so specific? Is the CAC approach inappropriate for detecting clusters regarding very specific events/concepts, even if they take up a lot of space (both textually and in the social consciousness)?
It's interesting to me how the authors' CAC approach was able to find a whole new (and significant) cluster, partisan taunting, in congressmen's press releases, but was not able to detect an important cluster, 9/11, in presidential addresses. Is it because the latter is so specific? Is the CAC approach inappropriate for detecting clusters regarding very specific events/concepts, even if they take up a lot of space (both textually and in the social consciousness)?