This directory presents frequency listings from three different types of
sources: "PAROLE", "Pressmaterial" and "Romaner".

PAROLE is a morphosyntactically tagged balanced corpus of appr. 20
mill. tokens  available through a graphical search interface at 
http://spraakbanken.gu.se/parole/ More infornation about PAROLE is to
be found at the same location. Unfortunately - in the case of the
PAROLE corpus - we don't have access to certain significant background
information,  which somewhat erodes the value of conclusions drawn from
that material.

Pressmaterial is an umbrella term for several collections of volumes of
Swedish daily newspapers. These materials correspond to the major part
of corpora available in http://spraakbanken.gu.se/konk/


Romaner currently contains statistics from two corpuses : 'Äl. sv.
rom.' (Older Swedish Novels) och  '1800-talsrom.'  (novels from the
19th century - a subselection from 'Äl. sv. rom.')

 In each and every subdirectory from here you will find "about"-files
describing the resource.