You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
This was part of Theresa Zieglers feature-list (and which is available in her Excel-Dashboard).
Multi-level-frequencies
Multi-level-frequencies are supported by NoSke - actually corpsum now uses the method freqs_ml which stands for multi-level.
This entails we nee to implement:
a way to actually specify a query for multi-level frequencies
-- either: add it to list of predefined searches to the dashboard (but then: what would be the list of necesssary+useful combinations? year+region and year+docsrc quite obviously. But what else? => linguists must tell!)
-- or: actually allow for a "free specification" of freq-parameters like in the NoSke)
a way to normalize the results: a crucial matter: NoSke nowadays provides correct normalisation but only for frequencies for 1 facette/text-type!
a way to present the results: 3D plots? In Theresas dashboard: a series of 2D plots. E.g. freqs for 5 x region + 20 x year: 5 plots (1 per region) with a extra plot per region with a 2D line plot "freqs over time"
The text was updated successfully, but these errors were encountered:
some scenarios can already be handled in the current corpsum version, but need a lot click work
the sketch engine itself does not handle normalisation for multi facetted searches
This was part of Theresa Zieglers feature-list (and which is available in her Excel-Dashboard).
Multi-level-frequencies
Multi-level-frequencies are supported by NoSke - actually corpsum now uses the method
freqs_ml
which stands for multi-level.This entails we nee to implement:
-- either: add it to list of predefined searches to the dashboard (but then: what would be the list of necesssary+useful combinations? year+region and year+docsrc quite obviously. But what else? => linguists must tell!)
-- or: actually allow for a "free specification" of freq-parameters like in the NoSke)
The text was updated successfully, but these errors were encountered: