Can anybody please advise good source(s) to read about the following?
How to extend standard statistical notions of similarity, distance,
influence, clustering, discrimination etc to sets of texts. I can
see many possible ways of doing this but would like advice on which
methods have been tried, which have good software available, and other
factors both practical and theoretical that I need to be aware of.
(The specific research project I have been asked to advise upon has
two main corpuses of texts: Corpus A relates to "research"; Corpus B
relates to "policy". Required is how to measure the impact of A upon
B.)
JOHN BIBBY
You may leave the list at any time by sending the command
SIGNOFF allstat
to [log in to unmask], leaving the subject line blank.
|