Universal POS, Detailed POS, NER, DEP

UPOS (Universal POS)

UPOS (Universal Part-of-Speech) tags are a core component of the Universal Dependencies (UD) project, designed to provide a standardized, fixed set of 17 categories that remain consistent across all human languages. Unlike language-specific systems (XPOS), which reflect the unique morphological intricacies of a single tongue, UPOS focuses on the functional role of a word. By stripping away language-specific "noise," UPOS allows researchers and developers to compare syntactic structures cross-linguistically and facilitates Cross-Lingual Transfer Learning—where an AI model trained on one language (like English) can apply its structural knowledge to another (like Romanian or Korean). It essentially serves as a "Lingua Franca" for computational linguistics, ensuring that a NOUN remains a NOUN whether the underlying grammar is agglutinative, fusional, or analytic.

Try our Russian UPOS tagging now.

UPOS Universal Part-of-Speech
Group Tag Meaning Example
Open Class ADJ Adjective большой, старый, зеленый, непонятный, первый
ADV Adverb очень, завтра, внизу, где, там
INTJ Interjection псст, ой, браво, привет
NOUN Noun (common) девушка, кот, дерево, воздух, красота
PROPN Proper Noun Мэри, Джон, Лондон, НАТО, HBO
VERB Verb беги, беги, беги, ешь, ел, ел
Closed Class ADP Adposition в, до, во время
AUX Auxiliary есть, сделал (сделал), будет (сделать), должен (сделать)
CONJ Conjunction и, или, но (устаревший тег)
CCONJ Coordinating Conjunction и, или, но
SCONJ Subordinating Conjunction если, пока, что
DET Determiner а, ан,
NUM Numeral 1 января 2017 года, семьдесят семь, MMXIV
PART Particle , а не
PRON Pronoun Я, ты, он, она, я, себя, кто-то
Other PUNCT Punctuation ., (, ), ?, ]
SYM Symbol $, %, +, −, :), 🐻
X Other / Foreign sfpksdpsxmsa, ..., foreign words
SPACE Space newlines, tabs, extra spaces

XPOS (Detailed POS)

XPOS (Language-Specific Part-of-Speech) tagging offers a much higher level of granularity than the broader UPOS (Universal Part-of-Speech) system. While UPOS provides a standardized set of labels designed to work consistently across every language—ensuring that a NOUN in English is treated similarly to a NOUN in XPOS preserves the unique "linguistic DNA" of a specific language. It is the engine behind complex morphological analysis, allowing a system to distinguish not just that a word is a "Verb," but specifically that it is a "Third-Person, Singular, Past Tense, Passive Voice" verb. By capturing the deep grammatical details that UPOS omits for the sake of universality, XPOS enables the creation of translation tools and parsers that understand the precise inflectional logic of a specific culture and tongue.

In French, Spanish, Portuguese, Danish, Norwegian, Russian, Hebrew, Catalan, Finnish, Sanskrit, Thai and Ukrainian, a separate fine-grained XPOS tagset is not defined. Instead, these languages utilize UPOS with specific granularities stored within Morphological Features.

Try our Russian XPOS tagging now.

General xpos Morphological Details
Group Category Label Meaning Example
Nominal Gender & Animacy Masc Masculine perro (dog)
Fem Feminine perra (female dog)
Neut Neuter ello (it/that)
Com Common estudiante
Hum Human persona, qui
Anim Animate (Living entity)
Inan Inanimate (Object)
Definite & Degree Def Definite le, la, el
Ind Indefinite un, une
Pos Positive degree bueno, bon
Cmp Comparative más, plus
Sup Superlative buenísimo
Nominal Number Sing Singular livre (book)
Plur Plural livres (books)
Nominal Case Nom Nominative yo, I
Acc Accusative me, lo
Dat Dative le, me
Gen Genitive (Possessive case)
Nominal NounType & NameType Class Classifier (NounType) ตัว (body/animal)
Giv Given Name (NameType) สมชาย (Somchai)
Sur Surname (NameType) ใจดี (Jaidee)
Geo Geographical (NameType) กรุงเทพฯ (Bangkok)
Nat Nationality (NameType) ไทย (Thai)
Com Company (NameType) กูเกิล (Google)
Verbal Mood & Aspect Ind Indicative yo hablo
Sub Subjunctive que yo hable
Imp Imperative ¡habla!
Cnd Conditional hablaría
Imp Imperfective hablaba
Perf Perfective hablé
Prog Progressive estoy hablando
Verbal Person & Politeness 1 First Person yo, nosotros
2 Second Person tú, vosotros
3 Third Person él, ella
Form Polite/Formal Usted, Vous
Infm Informal tú, toi
Verbal Tense Pres Present mange, eat
Past Past mangé, ate
Fut Future mangerai
Verbal VerbForm & Voice Fin Finite il court
Inf Infinitive courir, to run
Part Participle vu, visto
Ger Gerund corriendo
Act Active Voice veo (I see)
Pass Passive Voice soy visto
Lexical NumType Card Cardinal uno, deux
Ord Ordinal primero, 1er
Mult Multiplicative doble, triple
PronType Prs Personal yo, je
Dem Demonstrative este, celui
Rel Relative que, qui
Int Interrogative ¿quién?, qui?
Lexical Polarity & Poss Neg Negative no, pas
Yes Possessive mio, sien
Yes Reflexive se, me, te
Lexical PartType (Particles) Enp Ending Particle ครับ (krab), ค่ะ (kha)
Res Response Particle ใช่ (chai / yes)
Int Interrogative Particle ไหม (mai / ?)
Special Other Yes Foreign Word software, ad-hoc
Yes Abbreviation etc., adj.
Special Word Formation Yes (Prefix) Nominalizing Prefix การ- (kan-), ความ- (khwam-)
Rdp (Echo) Reduplicative เด็กๆ (dek-dek)

Dependency

The DEP (Syntactic Dependency) refers to the specific grammatical relationship between a "child" token and its "head" (parent) token. While primary labels (like nsubj or obj) describe the basic structure, attachments starting with a colon (:) provide fine-grained sub-type information. For instance, while nsubj identifies a subject, :pass refines this to show the subject is being acted upon (Passive Voice). Similarly, :nn (Noun Compound) or :assmod (Associative Modifier) help the parser distinguish between simple modifiers and complex ownership or compound relationships, allowing for a much deeper "logical" understanding of the sentence.

DEP Full Syntactic Dependency Labels
Category Label Meaning Example (Token in bold)
Core Arguments nsubj Nominal subject Илон ест.
csubj Clausal subject То, что он сделал, было неправильно.
obj Direct object Я вижу луну.
iobj Indirect object Она сделала мне подарок.
ccomp Clausal complement (finite) Он сказал, что устал.
xcomp Open clausal complement Я хочу пойти.
Non-Core Dependents obl Oblique nominal Он сел на стул.
vocative Vocative Джон, иди сюда!
expl Expletive Там есть кот.
dislocated Dislocated element Этот человек, я его знаю.
advcl Adverbial clause modifier Я ушел после его прибытия.
advmod Adverbial modifier Беги быстро.
discourse Discourse element Ну, я не уверен.
aux Auxiliary Я могу видеть.
cop Copula Она счастлива.
mark Subordinating marker Я знаю, что вы знаете.
Nominal Dependents nmod Nominal modifier Дверь автомобиля.
appos Appositional modifier Сэм, мой друг.
nummod Numeric modifier Семь дней.
acl Adjectival clause План на победу.
amod Adjectival modifier голубое небо.
det Determiner Конец.
case Case marking Король Франции.
fixed Fixed multiword expression Несмотря на это.
flat Flat multiword name Нью-Йорк Город.
compound Compound noun Телефонная будка.
list List element Телефон, ключи, кошелек.
Coordination conj Conjunct Хлеб и масло.
cc Coordinating conjunction Хлеб и масло.
Special Labels aux:pass Passive auxiliary Оно было украдено.
punct Punctuation Привет!
dep Unspecified dependency (Используется для неизвестных ссылок)
ROOT Root of the sentence Я съел обед.

Common Dependency Attachments (Sub-labels)
Attachment Full Name Explanation Example
:pass Passive Indicates a relationship in a passive voice construction. nsubj:pass (окно было разбито)
:nn Noun Compound Indicates that a noun is modifying another noun in a compound structure. соединение:nn (зарядное устройство телефона)
:prep Prepositional Refines a modifier governed specifically by a preposition. nmod:prep (Кошка на коврике)
:assmod Associative Modifier Common in Romanian/Baltic languages; shows nouns modifying other nouns. nmod:assmod (Машина моего отца)
:poss Possessive Indicates ownership or a possessive relationship. nmod:poss (Моя собака, Шляпа Джона)
:relcl Relative Clause Identifies a clause that modifies a noun phrase. acl:relcl (Книга, которую я прочитал)
:tmod Temporal Modifier A modifier specifically describing time or duration. nmod:tmod (я уезжаю вторник)
:prt Particle Used for phrasal verb particles. compound:prt (отключить включить, выключить выключить)
:rcomp Relative Complement Used for complements of relative clauses (common in Dutch). advcl:rcomp (Человек, который ушел)
:flat Flat Modifier Used for multi-word expressions that don't have a clear internal head. квартира:имя (Президент Обама)

Named Entity Recognition

NER (Named Entity Recognition) is a Natural Language Processing (NLP) task that automatically identifies and categorizes key information (entities) in a text into predefined classes. In spaCy, the statistical model "looks" at the context of a word to determine if it refers to a person, an organization, a monetary value, or a specific date. This is crucial for extracting structured data from unstructured text, such as finding all the company names mentioned in a news article or identifying the dates of events in a history book.

Comparison Note: GPE vs. LOC
Determining whether a place is a GPE or a LOC depends on its political nature:
GPE (Geopolitical Entity): If the location has a government, specific laws, or human-defined administrative borders, it is labeled as a GPE. Examples include Seoul, Germany, the United Kingdom, and California.
LOC (Location): If the place is a natural physical feature or a broad geographic region without a singular governing body, it is labeled as a LOC. Examples include the Alps, the Pacific Ocean, the Middle East, and Mount Everest.

NER Named Entity Recognition
Label Meaning Example
🌍 GPE Geopolitical entity (countries, cities, states) США, Нью-Йорк, Франция, Калифорния
🏔️ LOC Non-political location (mountains, rivers) Тихий океан, Эверест, Альпы
🏢 FAC Facility (buildings, airports, highways) Мост Золотые Ворота, аэропорт имени Джона Кеннеди, Бурдж-Халифа
👤 PERSON People (real or fictional) Илон Маск, Гарри Поттер, Алан Тьюринг
🚩 NORP Nationalities, religious or political groups Американцы, буддисты, демократы, японцы
🏢 ORG Organizations (companies, institutions) Google, ООН, Apple, FIFA
📅 DATE Absolute or relative dates 4 июля 2026 г., вчера, на следующей неделе
⌚ TIME Times smaller than a day 9:30 утра, закат, десять минут
🎊 EVENT Named events (wars, festivals) Вторая мировая война, Коачелла, Олимпийские игры
💰 MONEY Monetary values, including unit 100 долларов США, 5 миллионов евро, 50 фунтов стерлингов
‱ PERCENT Percentage, including "%" 20%, восемьдесят процентов, 0,5%
⚖️ QUANTITY Measurements (weight, distance) 5 км, 100 фунтов, 30 квадратных метров
🔢 ORDINAL "First", "second", etc. первый, 2-й, девятый
🔢 CARDINAL Numbers not classified elsewhere 10, тысяча, три
📦 PRODUCT Objects, vehicles, foods, etc. (not services) iPhone, Tesla Model S, Coca-Cola
🎨 WORK_OF_ART Titles of books, songs, etc. Мона Лиза, Богемская рапсодия, Гамлет
📜 LAW Named legal documents Конституция, Версальский договор
🗣️ LANGUAGE Named languages Английский, Python, Мандарин

Пример НЛП (NLP Example)

Если мы обработаем фразу «Google базируется в Калифорнии», слои будут выглядеть так:

Лемма: "Google", "be", "base", "in", "California"
УПОС: "PROPN(Proper Noun)", "AUX(Auxiliary)", "VERB(Verb)", "ADP(Adposition)", "PROPN(Proper Noun)"
XPOS: "NNP(Proper noun, singular)", "VBZ(Verb, 3rd person singular present)", "VBN(Verb, past participle)", "IN(Preposition or subordinating conjunction)", "NNP(Proper noun, singular)"
DEP: "Google" — это nsubj (именное подлежащее) глагола "based", который является Root (коренем предложения).
NER: «Google» — это 🏢 ORG (организация), «Калифорния» — это 🌍 GPE (геополитическая организация).

Part-of-Speech for Main Languages

Arabic - Catalan - Chinese - Classical Chinese - Croatian - Danish - Dutch - English - Filipino - Finnish - French - German - Greek - Hebrew - Hindi - Italian - Indonesian - Japanese - Korean - Latin - Lithuanian - Macedonian - Norwegian - Polish - Portuguese - Romanian - Russian - Slovenian - Sanskrit - Spanish - Swedish - Tamil - Thai - Ukrainian - Vietnamese

  • Home
  • Translators
  • Dictionaries
  • Grammars
  • Keyboards
  • Facebook

    © Stars21 - All Rights Reserved