Kanones.jl

A julia package implementing the kanónes system for building Greek morphological parsers.

For a succinct outline of the technical process, see the project wiki.

Current coverage

Kanones.jl is being developed with test datasets in three distinct orthographies: standard literary Greek as it appears in printed editions, an extended literary Greek orthography including symbols found in some manuscripts, and an orthography for the archaic and classical Attic alphabet used for public documents in Athens before 403 BCE.

Kanones.jl recognizes nine analytical types, defined by the kind of analysis a parse returns. The following table tracks coverage so far of analyzing and generating forms for each analytical type in both literary Greek and Attic Greek orthographies.

Analytical typemorphologyanalyze literary Greek regular formanalyze literary Greek irregular formgenerate literary Greekgenerate literary Greek irregularanalyze Attic regular formanalyze Attic irregular formgenerate Attic
uninflected tokenpart of speech✔️N/A✔️N/A✔️N/A✔️
noungender, case, number (gender inherent)✔️✔️✔️✔️✔️✔️✔️
pronoungender, case, number (gender not inherent)N/A✔️N/A
finite verbperson, number, tense, mood, voice✔️✔️
infinitivetense, voice✔️✔️
participletense, voice, gender, case, number✔️
verbal adjectivegender, case, number✔️N/AN/A
adjectivegender, case, number, degree✔️
adverbdegreeN/AN/A