A strange result I found was that nearly all of the incorrect classifications of Middle English were in a specific file of A Treatise on the Astrolabe by Chaucer. The other texts by Chaucer had much greater accuracy than the average shown in the graphs.
Monday, July 20, 2009
There's something about "A Treatise on the Astrolabe"
From this paper on natural language classification by character frequency:
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment