[ Skip to the content ]

Institute of Formal and Applied Linguistics Wiki


[ Back to the navigation ]

This is an old revision of the document!


Table of Contents

DZ Interset Features and Values

pos

Part of speech. Pronouns, determiners, predeterminers and articles are roofed by nouns and adjectives, and distinguished by values of other features.

Value Description
noun noun
adj adjective
num numeral, number
verb verb
adv adverb
prep preposition, postposition or circumposition (together also: adposition)
conj conjunction
part particle
int interjection
punc punctuation or symbol

subpos

Detailed part of speech.

Value Main pos Description
prop noun proper noun (“George”, “Bush”, “Paris”)
class noun classifier (measure word) between number and counted noun, e.g. Chinese “個 gè”
pdt adj predeterminer (adjectival word that can stand before an article, such as “all” in “all the flowers”)
det adj determiner (function word modifying a noun phrase: English “this”, “that”); regarded indefinite/demonstrative pronoun in some tagsets; includes articles (see below) in some tagsets
art adj article, i.e. determiner bearing only the feature of definiteness or indefinitess and nothing more (English “a”, “an”, “the”, German “der”, “die”, “das”, Portuguese “um”, “uma”, “o”, “a”, “os”, “as”)
aux verb, part auxiliary verb used to construct complex verb forms (Czech “být”, English “have”, “will”)
cop verb copula verb (Czech “být”)
mod verb modal verb (German “dürfen”, “können”, “mögen”, “müssen”, “sollen”, “wollen”, “wissen”; Czech “muset”, “mít”, “moci”, “smět”, “umět”, “chtít”; English “must”, “can”, “shall”); note that adverbs and particles have their own mod subpos
verbconj verb finite verb with the enclitic “-ť” (Czech “neboť” = “because”)
man adv adverb of manner
loc adv adverb of location
tim adv adverb of time
deg adv adverb of quantity or degree
cau adv adverb of cause (“why”)
mod adv, part modal particle (Bulgarian “май” = “possibly”, “нека” = “let”; Czech “ať”, “kéž”, “nechť”) or adverb of modal nature (Bulgarian “апропо”); note that verbs have their own mod subpos
ex adv existential “there” in English
voc prep vocalized preposition (Czech “ve” as opposed to base form “v”)
post prep postposition (German “entlang” in “der Strasse entlang”)
circ prep circumposition (German “von … an” in “von dieser Stelle an”)
preppron prep preposition and pronoun in one word (Czech “proň” = “pro něj”, “nač” = “na co”)
comprep prep first part of compound preposition (Czech “nehledě na”, “vzhledem k”)
coor conj coordinating conjunction
sub conj subordinating conjunction
comp conj comparing conjunction (German “wie”, “als”)
emp part particle of emphasis (Bulgarian “даже” = “even”)
res part particle of response (“yes”, “no”)
inf part infinitive marker (English “to”, German “zu”, Danish “at”, Swedish “att”). Sometimes tagged as particle, sometimes as conjunction, sometimes has its own part of speech.
vbp part separated verb prefix (German “vor” in “stellen Sie sich vor”); analogical verbal particles in English?

prontype

This is a new (September 2007) feature applied first to the Bulgarian CoNLL tag set. It takes over the pronoun classification that has been so far kept in the definiteness feature. See the brainstorming section for further details on lexical and morphological definiteness.

Although it reads as “pronoun type” (and we use the word “pronoun” for simplicity), it is also applied to words that are usually not considered pronouns, such as interrogative/indefinite adverbs (where, there, when, then, how, why).

Value Description
Empty value means that this is not a pronoun but a real noun, adjective, adverb etc. This will be more useful once we completely remove the pronoun part of speech, distribute the pronouns to other syntactically similar parts of speech and set their prontype.
prs Personal or possessive pronoun. Possessives are recognizable by the value of their poss feature.
rcp Reciprocal pronoun (German “einander”, Danish “hinanden”). Similar to personal pronouns but occurs as special case in object position.
int Interrogative pronoun (“who”, “what”, “which”).
rel Relative pronoun. Many interrogative pronouns in many languages can also be used as relative pronouns. However, in some languages there are pronouns that fall in one of the categories but not both (Czech “jenž” is only relative; in Bulgarian, relatives are completely separated from interrogatives). For words that can be both interrogative and relative, “int” is the default value.
dem Demonstrative pronoun (“this”, “that”). Being a demonstrative pronoun is not the same as being definite (definiteness=def), although the two feature-values are similar.
neg Negative pronoun (“nobody, nothing, none”). This is not the same as the negativeness feature. Unlike e.g. negative and positive adjectives or verbs, negative pronouns are not complements of some “positive” pronouns. Instead, they usually correspond to zero, nothing.
ind Indefinite pronoun (“somebody”, “something”, “anybody”, “anything”). Being an indefinite pronoun is not the same as being morphologically indefinite (definiteness=ind). For instance, in Bulgarian there are morphologically definite lexically indefinite pronouns (“едната”, “едното”, “едните”, “нещата”). In some languages, we could subclassify the indefinite pronouns into “few” (“málokdo”), “ind” (“někdo”), “mny” (“leckdo”), “any” (“kdokoli” - anybody you pick but you pick only one, not all at once; this is the difference from the totality pronouns)
tot Totality pronoun (“everybody”, “everything”)

numtype

Subclasses of numerals. See also prontype (interrogative and indefinite numerals set it), numform and numvalue.

Value Description
card cardinal number
ord ordinal number
mult multiplier number (“five times”)
frac fraction (“one fifth”)
gen generic numeral (“twofold”, Czech “jedny”, “čtvery”, “čtvero”, “čtverý”)

numform

Is a number expressed by a word or by digits? Depending on tokenization and tagging scheme, this feature may be orthogonal to the distinction between cardinals and ordinals.

Value Description
word numeral word (“fourteen”)
digit number written using digits (“14”)
roman number written using Roman numerals (“XIV”)

numvalue

Low-value numerals display special behavior in some languages (e.g. Czech). This feature helps distinguish them.

Value Description
1 numeric value 1 (Czech “jeden”, “první”)
2 numeric value 2 (Czech “druhý”, “dvojí”, “dvojnásob”, “dva”, “nadvakrát”, “oba”, “obojí”)
3 numeric value 3 or 4 (Czech “čtvrtý”, “čtyři”, “potřetí”, “tři”, “třetí”, “třikrát”)

punctype

Punctuation type.

Value Description
peri period at the end of sentence; in Penn tagset, includes question and exclamation
qest question mark
excl exclamation mark
quot quotation marks
brck bracket
comm comma
colo colon; in Penn tagset, “:” is in fact tag for generic other punctuation
semi semicolon
dash dash
symb symbol
root artificial sentence root node, beginning of sentence

puncside

Distinguishes between initial and final form of pairwise punctuation (brackets, quotation marks). Note that “initial” and “final” are better terms than “left” and “right”. The latter would be confusing in languages writing from right to left, like Arabic.

Value Description
ini initial (left bracket in English texts)
fin final (right bracket in English texts)

synpos

Does the pronoun or numeral behave syntactically as a noun, adjective, or adverb?

Value Description
subst substantive (like a noun)
attr attributive (like an adjective)
adv adverbial (like an adverb)
pred predicative adjective

poss

Is this a possessive adjective or pronoun?

Value Description
poss possessive

reflex

Is this a reflexive pronoun?

Value Description
reflex reflexive

negativeness

Distinguishes also negative pronouns like “nothing.”

Value Description
pos positive, affirmative
neg negative

definiteness

See also the prontype feature.

Value Description
ind indefinite
def definite
red reduced: used in construct state in Arabic. If two nouns are in genitive relation, the first one has “reduced definiteness,” the second is the genitive.

foreign

Value Description
foreign foreign word (not a loan word but a citation in a foreign language — e.g., the title of a foreign book)

gender

Value Description
masc masculine
fem feminine
com common, utrum
neut neuter

possgender

Possgender is possessor's gender.

Value Description
masc masculine
fem feminine
com common, utrum
neut neuter

animateness

Value Description
anim animate
inan inanimate

number

Value Description
sing singular
dual dual
plu plural

possnumber

Possnumber is possessor's number.

Value Description
sing singular
dual dual
plu plural

case

Value Description
nom nominative
gen genitive
dat dative
acc accusative or oblique
voc vocative
loc locative
ins instrumental

prepcase

Personal pronouns in some languages have different forms depending on whether they are objects of prepositions or not. For instance, Czech “on” (he) without prepositions has the forms jemu/DAT, jeho/ACC, jím/INS, while with a preposition it is němu/DAT, něho/ACC, ním/INS. Similarly, Portuguese pronouns in prepositional oblique case take forms different from oblique pronouns serving as direct objects of verbs: eu/NOM (I), me/ACC (give me that), mim/PREP-ACC (come to me).

Default empty value means that the word form is neutral w.r.t. a prepositional head.

Value Description
npr this form must not be used after a preposition (Czech “jemu”)
pre this form must be used after a preposition (Czech “němu”)

degree

Degree of comparison.

Value Description
pos positive, first degree (note that although this degree is traditionally called “positive”, negative properties can be compared, too)
comp comparative, second degree
sup superlative, third degree
abs absolute superlative

person

Value Description
1 first (I, we)
2 second (you)
3 third (he, she, it, they)

politeness

Value Description
inf informal (Czech “ty/vy”, German “du/ihr”, Spanish “tú/vosotros”)
pol polite (Czech “vy”, German “Sie”, Spanish “usted”)

subcat

There are tag sets (e.g. Bulgarian CoNLL) that classify verbs as intransitive or transitive. It turns out that a Bulgarian verb can have set both features type=aux && trans=t. That is why we cannot mix transitivity and auxiliarity in subpos.

Value Description
intr intransitive verb
tran transitive verb

verbform

Value Description
fin finite
inf infinitive
sup supine (with motion verbs: “go do something”; infinitive used in languages where there is no supine)
part participle (present (“doing”), past (“done”), passive (Czech “udělán” distinguished from adjective “udělaný” by variant=short)), gerundive
trans transgressive, adverbial participle (modifies other verbs, behaves like adverb; Czech present “dělaje”, past “udělav”)
ger gerund (verbal noun)

mood

Value Description
ind indicative
imp imperative
cnd conditional
sub subjunctive (conjunctive) (spojovací)
jus jussive (přací)

tense

Value Description
past past
pres present
fut future

subtense

Finer classification of tenses, may not be available in all languages. (And in many languages, these tenses are built using auxiliaries, rather than special morphemes.) Having these separated from the main past-present-future distinction allows that drivers need not check for aorist/imperfect, if they know just one past tense.

Note that, unfortunately, imperfect tense is not always the same as past tense + imperfective aspect. For instance, in Bulgarian, there is lexical aspect, inherent in verb meaning, and grammatical aspect, which does not necessarily always match the lexical one. In main clauses, imperfective verbs can have imperfect tense and perfective verbs have perfect tense. However, both rules can be violated in embedded clauses. Aorist is aspect-neutral and can freely appear with both imperfective and perfective verbs.

Value Description
aor aorist
imp imperfect
pqp pluperfect

aspect

Value Description
imp imperfect
perf perfect

voice

Value Description
act active
pass passive

abbr

Is this an abbreviation?

Value Description
abbr abbreviation

hyph

Is this a part of a hyphenated compound?

Value Description
hyph hyphenated prefix (“anglo-” in “anglo-saxon”)

style

Value Description
arch archaic, obsolete, rare
form formal, literary
norm normal, neutral
coll colloquial

typo

Indicates bad spelling, grammatical error etc. Does not say how the correct form looks like. Must be separated from the style feature because there can be typo in archaic as well as colloquial form.

Value Description
typo typo, bad spelling, error

variant

Allows for distinguishing between word forms that otherwise would share values of all features.

Value Description
short short form
long long form
0 variant form 0
1 variant form 1
2 variant form 2
3 variant form 3
4 variant form 4
5 variant form 5
6 variant form 6
7 variant form 7
8 variant form 8
9 variant form 9

tagset, other

The tagset feature identifies the source tag set driver. Value should be identical to the name of the driver that filled the feature values. (The recommended format is same as in the use statement, without the tagset:: prefix, e.g. “de::stts”. Works together with the “other” feature.

Feature “other”
Any value or reference to array or hash is allowed. Serves to preserve information if the decoding driver happens to be the one who did the encoding. No other driver should expect anything meaningful here.
Only information that cannot be stored in other features should be stored here.
The apparently easiest approach — to store the complete original tag — will not work if the user needs to change feature values between decode() and encode().


[ Back to the navigation ] [ Back to the content ]