Skip to content

Correcting some unknown Part of Speech tags #21

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Oct 5, 2018

Conversation

todd-cook
Copy link
Contributor

@todd-cook todd-cook commented Oct 5, 2018

Correcting some unknown Part of Speech tags

Some words are tagged with an unknown part of speech tag: "---------"
But many of the unknowns can be corrected and reasonably resolved.

This PR aims to improve the POS tag quality for the latin corpus.

People processing these tag sets will likely want to handle or drop the rest of the unknowns,
or unknowns between punctuation (i.e. alternate readings) as indicated below.

Corrections:

  1. An abbreviated noun in apposition to a name is attracted to the same declension (see Greenough, section 282, Apposition)

e.g.
<word id="5" form="L" lemma="L1" postag="---------" relation="ATR" head="7"/>
<word id="6" form="." lemma="PERIOD1" postag="u--------" relation="AuxG" head="5"/>
<word id="7" form="Sullae" lemma="Sulla1" postag="n-s---mg-" relation="ATR" head="4"/>
2. Obvious corruptions in the sequence of tags in the text, e.g.:
<word id="5" form="po" lemma="other" postag="---------" relation="XSEG" head="6"/>
<word id="6" form="puli" lemma="populus" postag="n-s---mg-" relation="ATR" head="4"/>
<word id="7" form="Romani" lemma="Romanus1" postag="a-s---mg-" relation="ATR" head="6"/>
or:
<word id="4" form="c" lemma="que1" postag="---------" relation="COORD" head="1"/>
<word id="5" form="ne" lemma="ne1" postag="d--------" relation="AuxZ" head="6"/>

...etc.

Items not corrected:

  1. Greek characters represented in Arabic fonts/transliterated Greek passages, e.g.:
    <word id="1" form="ma" lemma="other" postag="---------" relation="AuxZ" head="3"/>
    <word id="2" form="/" lemma="SLASH1" postag="u--------" relation="AuxG" head="1"/>
    <word id="3" form="deia" lemma="other" postag="---------" relation="AuxY_ExD_OBJ" head="0"/>
    <word id="4" form="perima" lemma="other" postag="---------" relation="AuxZ" head="6"/>
  2. Bracketed editorial items; partial words, e.g.:
    <word id="21" form="[" lemma="bracket1" postag="u--------" relation="AuxG" head="22"/>
    <word id="22" form="dict" lemma="other" postag="---------" relation="XSEG" head="24"/>
    <word id="23" form="]" lemma="bracket1" postag="u--------" relation="AuxG" head="22"/>
    <word id="24" form="ita" lemma="ita1" postag="d--------" relation="PRED_CO" head="19"/>

Texts used:

Source citations; corrections against Perseus texts
Sallust’s Bellum Catilinae (2007, 2nd ed.) by J. T. Ramsey
Sallust’s Bellum Catilinae, 2018, Geoffrey Steadman
Petronius Cena Trimalchionis, 2018, Geoffrey Steadman

Some words are tagged with an unknown part of speech tag: "---------"
But many of the unknowns can be corrected and reasonably resolved.
@gcelano gcelano merged commit 74a6752 into PerseusDL:master Oct 5, 2018
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants