Release 0.8.2

Minor fix release (server migration in Nijmegen), updated tests

Release 0.8.1

This release hopes to fix the issue of Alpino failing by time-out on very long sentences (a2ed0f5).

Release 0.8

New features in this release:

• Added features with corrected word frequencies on word/sentence level (bd0bfa6, c41aef1)
• Made sure both Wopr instances run in parallel (7de9f9a)

Release 0.7

New features in this release:

• Added features to calculate the total number of propositions (cd50918)
• Added features that remove a specific set of adverbs from the content counts (8f203ba)
• Updated download URL for bigdata (276497b)

The following bugs were squashed:

• Fixing Conn_d and Conn_dz on paragraph and document level (fa3e40f)
• Using the lowercase word for adverb lookup (fd98261)

Release 0.6

New features in this release:

• Added subtype for Adverbs, only limited set of adverbs now counts as content words (ae4bc70)
• Added features for totals of connectives (5d50f50)
• Added Belgian prevalence data, created a switch on the CLAM front-end to choose between the two datasets (6217706, 7d553de)
• Removed the rarity level as a parameter on the front-end (2bc1377)

Release 0.5

Two new features in this release:

• Functionality to retrieve prevalence scores (c04913c)
• Eigen_classificatie feature on sentence, paragraph and document level (15f83ba)

Other than that, some small clean-up actions.

Release 0.4.1

New conventions for frog/ucto/tscan data: installing in share/instead of etc/.

Release 0.4

First release via GitHub. Uses up-to-date versions of libfolia and frog, as well as fixes some issues with the frequrency count handling. We plan to follow a more stringent/semantic release protocol from now on :-)