I have used xiaoz's detagging tool to remove everything other than the orignal texts and transcripts of the BNC's raw materials. it's really a great software! But when processing the documents in txt form by wordsmith, i can get correct sd. T/T ratio and other statistics. the problem is sentence length and sd. sentence length are always above 200, sometimes even 300-odd. i checked the txt, there is nothing wrong with the punctuation. how to solve the problem? what's wrong with my work? many thx for help
Last edited by a moderator: