Norsk

 • Read about OBT

 • History

 • Evaluation

 • Tagset

 • OBT in use

 • Publications

 • Download

 • Contact

Punctuation marks

                                                                                                                                      <— Back to tagset

The tagger marks sentence boundaries with clb (for example with commas and conjunctions). Complete sentence boundaries are marked with <<< .

Quotation marks and dots are put together with the word to the left or right if the character is assumed to be a part of the word, eg. when the dot is part of an abbreviation: A. subst prop fork, or when a single word only is in quotes: <"exit">. (When multiple words are framed by quotation marks, quotation marks are separated as a separate character <">.)

The tagger also tries to identify headlines, and gives them a complete sentence boundary tag plus a character | meaning headline:

<word>Været</word> 'the weather'
"<været>"
      "vær" subst appell nøyt be ent
<word>blir</word> 'gets'
"<blir>"
     "bli" verb pres i2 tr5 a5 pa4/til pr1 pr2 <aux1/perf_part>
<word>bedre</word> 'better'
"<bedre>"
     "god" adj komp
"<|>" 'headline'
"$|" clb <overskrift> <<< ' sentence boundary'

Note that all uppercase letters are converted to lowercase, but that the original text is retained between the tags <word> </ word>.