Characteristics of the Polish language in AX Semantics

Fundamentals

Polish has three genders for nouns: masculine, feminine and neuter. Masculine nouns also differentiate between personal, animate and inanimate. Furthermore, there are two numbers: singular and plural. Polish has seven cases for nouns: nominative, accusative, genitive, dative, instrumental, locative, and vocative.

grammatical namevaluesexamples
gendermasculine inanimatestary port
(the old port)
masculine animatestary pies
(the old dog)
masculine personal groupstary człowiek
(the old man)
femininestara kobieta
(the old woman)
neuterstare krzesło
(the old chair)
------
numbersingularzerwony dom
(a red house)
pluraldwa duże krzesła
(two big chairs)
------
cases (noun)nominativepies
(the dog)
accusativeWidzę psa.
(I see the dog)
genitivedzwonek psa
(the dog's bell)
dativeDaję psu jego piłkę.
(I give the ball to the dog.)
instrumentalNiewidomy chodzi z psem.
(The blind man walks with a dog.)
locativeKot siedzi na psie.
(The cat sits on the dog.)
vocativeTo było dobre, psie!
(This was good, dog.)
------
verb tensespresentOn pisze
(He writes)
pastOn pisał (He wrote)/
Ona pisała (She wrote)
futureOn będzie pisał (He will write)

The standard order of a noun phrase in Polish is the following: preposition + determiner + numeral + adjective + noun.

See for example:

o      tych   trzech  popularnych  książkach
about  these  three   popular      books[pl,loc]
PREP   DET    NUM     ADJ          NOUN
"about these three popular books"

Lexicon

Nouns

For Polish nouns the lexicon needs to encode gender, animacy, preposition changes and case changes. If the lexicon entry is missing, the NLG platform tries to find the most probable gender based on heuristics. The case forms should be added to the lexicon if they are not regular.

Lexicon entries for nouns may also be necessary for inflecting determiners, adjectives, numerals and pronouns correctly. They are omitted, if a lexicon entry is required, but missing.

Examples

The basic lexicon entry for dom (house) contains:

SingularPlural
gendermale inanimatemale inanimate
Nominativedomdomy
Genitivedomudomów
Dativedomowidomom
Accusativedomdomy
Instrumentaldomemdomami
Locativedomudomach
Vocativedomiedomy

Additionally, the lexicon entry for Seychellen (Seychelles) contains:

  • gender, case, and number (like above inflection table)
  • replace preposition do with na in accusative

note

If you need lexicon entries for countries, write to the support about that and you will get them for Polish with automatic handling of determiners.

Adjectives

Polish adjectives agree with the noun in gender/animacy, number and case. For adjectives that inflect irregularly lexicon entries may need to be added.

Verbs

The most common verbs are encoded in our software. If a verb inflects the wrong way, you should add it to the lexicon.

Please note that the future tense can be formed by taking the future tense form of the verb być (to be) and putting the past tense form of the main verb in a separate container after it.

Container settings

Determiner

The AX NLG platform supports the following determiners in Polish: demonstrative, every, not, and possessives. Polish determiners exist as independent modifiers, for example:

Twoja gazeta jest przestarzała.
(Your newspaper is outdated.)

Ja czytam v tę gazetę.
(I’m reading this newspaper.)

Numerals

The noun will automatically agree with the numeral number when a numeral variable is used. Four types of numerals are possible on the AX NLG platform: cardinal, cardinal as digit, ordinal, and ordinal as digit.

cardinalordinal
textDziewięć myszy w moim pokoju.
(Nine mice in my room.)
Dziewiąty dzień w szkole.
(The ninth day at school.)
digit9 myszy w moim pokoju.
(9 mice in my room.)
Dziewiąty dzień w szkole.
(The 9th day at school.)

For Polish, ordinal numerals are written out until 99, otherwise (above 99) the output is in digit form. Cardinal numerals are written out until 100, otherwise (above 100) the output is in digit form. Take cardinal numerals for example,

sto samochodów
(one hundred cars)
101 samochodów
(101 cars)

Preposition

If users configure prepositions in the container, they are automatically adapted, if phonetic assimilation needs to happen. The prepositions w and z change to we and ze before words starting with a consonant. For example:

we Wrocławiu
(in Wrocław)

w Austrii
(in Austria)

Examples

Preposition switch

In the sentence

Jechali z Paryża do Australii.
(They were traveling from Paris to Australia.)

no article is added to Australii and the preposition do isn't changed.

On our platform, the settings for the container Australii are: preposition="do" and case="genitive".

For the same sentence but with another country (i.e. Seszele (Seychelles)) the platform settings are the same, but the lexical information is changing the results:

Jechali z Paryża na Seszele.
(They were traveling from Paris to the seychelles.)

Observing the two examples above, the preposition is changed from do to na and the case is switched to accusative.