previous class main page next class

Språkstatistik HT96:13

The ninth and the tenth lecture will be an introduction to simple statistical methods which can be applied to text corpora. Topics we will deal with are n-grams, mutual information, t-scores, n-gram language models and entropy. The statistical methods will be accompanied by sets of UNIX commands for implementing them.


Last update: August 19, 1996. erikt@stp.ling.uu.se