Figure 6. Semantic similarity between transcripts and texts in MT and HT.

Note that a value of 0 means purely random association and a value of 1 means identity. Thus, lighter hue means more similar semantic content. Here, this is visible in the bottom half of the left-hand figure, which shows that transcripts were more similar to Dostoevsky selections than to each other. Note that, unlike in MT, HT transcripts in aggregate semantically resemble each other more than they resemble the Chiang texts. MT=Michaelmas Term; HT=Hilary Term.