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 Polish UPOS tagging now.
| Group | Tag | Meaning | Example |
|---|---|---|---|
| Open Class | ADJ | Adjective | duży, stary, zielony, niezrozumiały, pierwszy |
| ADV | Adverb | bardzo, jutro, w dół, gdzie, tam | |
| INTJ | Interjection | pst, ałć, brawo, cześć | |
| NOUN | Noun (common) | dziewczyna, kot, drzewo, powietrze, piękno | |
| PROPN | Proper Noun | Maria, Jan, Londyn, NATO, HBO | |
| VERB | Verb | biegać, biega, biegający, jeść, jadł, zjedzony | |
| Closed Class | ADP | Adposition | w, do, podczas |
| AUX | Auxiliary | jest, zrobił, zrobi, powinien | |
| CONJ | Conjunction | i, lub, ale (stary tag) | |
| CCONJ | Coordinating Conjunction | i, lub, ale | |
| SCONJ | Subordinating Conjunction | jeśli, podczas gdy, że | |
| DET | Determiner | —, —, — | |
| NUM | Numeral | 1, 2017, jeden, siedemdziesiąt siedem, MMXIV | |
| PART | Particle | —, nie | |
| PRON | Pronoun | ja, ty, on, ona, ja sam, oni sami, ktoś | |
| Other | PUNCT | Punctuation | ., (, ), ?, ] |
| SYM | Symbol | $, %, +, −, :), 🐻 | |
| X | Other / Foreign | sfpksdpsxmsa, ..., foreign words | |
| SPACE | Space | newlines, tabs, extra spaces |
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.
Polish XPOS tags follow the NKJP (Narodowy Korpus Języka Polskiego) tagset logic. These tags identify the core morphological category of a word. For example, SUBST represents a substantive (noun), while PRAET identifies the past tense (praeterite). Because Polish verbs have highly distinct forms for present/future (FIN), past (PRAET), and various participles (PACT, PCON), the XPOS label is vital for determining the "tense-aspect" structure of a sentence.
Try our Polish XPOS tagging now.
| Category | Tag | Meaning | Example |
|---|---|---|---|
| Nouns & Pronouns | SUBST | Substantive (Common Noun) | dom (house), kot (cat) |
| PPRON12 | Personal Pronoun (1st/2nd Person) | ja (I), ty (you), my (we) | |
| PPRON3 | Personal Pronoun (3rd Person) | on (he), ona (she), oni (they) | |
| SIEBIE | Reflexive Pronoun | siebie, się | |
| Verbs & Participles | FIN | Finite Verb (Present/Future) | robi (does), kupi (will buy) |
| PRAET | Past Tense (Praeterite) | robił (did), biegli (ran) | |
| INF | Infinitive | robić (to do), być (to be) | |
| IMPT | Imperative | rób (do!), czekaj (wait!) | |
| IMPS | Impersonal Past | zrobiono (it was done), bito | |
| PCON | Contemporary Adverbial Participle | idąc (while walking), robiąc | |
| PAN | Anterior Adverbial Participle | zrobiwszy (having done) | |
| GER | Gerund (Verbal Noun) | pływanie (swimming), picie | |
| Modifiers | ADJ | Adjective | dobry (good), wielka (great) |
| ADJP | Post-prepositional Adjective | polsku (as in "po polsku") | |
| ADV | Adverb | dobrze (well), szybko (fast) | |
| PACT / PPAS | Adjectival Participle (Active/Passive) | palący (smoking) / kupiony | |
| Function Words | PREP | Preposition | w (in), na (on), do (to) |
| CONJ | Coordinating Conjunction | i (and), ale (but), lub (or) | |
| COMP | Subordinating Conjunction | że (that), bo (because), gdy | |
| QUB | Particle (Quasipartikel) | czy (question marker), nie (not) | |
| NUM | Numeral | pięć (five), dwaj (two) | |
| INTERJ | Interjection | halo, ach, ojej | |
| Others | AGLT | Agglutinate (Clitic endings) | -em (as in "robił+em") |
| INTERP | Punctuation | . , ? ! : | |
| XXX | Foreign word / Unrecognized | iPhone, Google |
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.
| Category | Label | Meaning | Example (Token in bold) |
|---|---|---|---|
| Core Arguments | nsubj | Nominal subject | Elon je. |
| csubj | Clausal subject | To, co zrobił, było złe. | |
| obj | Direct object | Widzę księżyc. | |
| iobj | Indirect object | Dała mi prezent. | |
| ccomp | Clausal complement (finite) | Powiedział, że jest zmęczony. | |
| xcomp | Open clausal complement | Chcę iść. | |
| Non-Core Dependents | obl | Oblique nominal | Usiadł na krześle. |
| vocative | Vocative | Janku, chodź tutaj! | |
| expl | Expletive | Tam jest kot. | |
| dislocated | Dislocated element | Znam tamtego człowieka. | |
| advcl | Adverbial clause modifier | Wyszedłem po tym, jak przyszedł. | |
| advmod | Adverbial modifier | Biegnij szybko. | |
| discourse | Discourse element | Cóż, nie jestem pewien. | |
| aux | Auxiliary | Mogę zobaczyć. | |
| cop | Copula | Ona jest szczęśliwa. | |
| mark | Subordinating marker | Wiem, że wiesz. | |
| Nominal Dependents | nmod | Nominal modifier | Drzwi samochodu. |
| appos | Appositional modifier | Sam, mój przyjaciel. | |
| nummod | Numeric modifier | Siedem dni. | |
| acl | Adjectival clause | Plan, aby wygrać. | |
| amod | Adjectival modifier | Niebieskie niebo. | |
| det | Determiner | Koniec. | |
| case | Case marking | Król Francji. | |
| fixed | Fixed multiword expression | Pomimo tego. | |
| flat | Flat multiword name | Miasto Nowy Jork. | |
| compound | Compound noun | Budka telefoniczna. | |
| list | List element | Telefon, klucze, portfel. | |
| Coordination | conj | Conjunct | Chleb i masło. |
| cc | Coordinating conjunction | Chleb i masło. | |
| Special Labels | aux:pass | Passive auxiliary | To zostało skradzione. |
| punct | Punctuation | Cześć! | |
| dep | Unspecified dependency | (Używane do nieznanych relacji) | |
| ROOT | Root of the sentence | Zjadłem obiad. |
| Attachment | Full Name | Explanation | Example |
|---|---|---|---|
| :pass | Passive | Indicates a relationship in a passive voice construction. | nsubj:pass (Okno zostało rozbite) |
| :nn | Noun Compound | Indicates that a noun is modifying another noun in a compound structure. | compound:nn (Ładowarka do telefonu) |
| :prep | Prepositional | Refines a modifier governed specifically by a preposition. | nmod:prep (Kot na macie) |
| :assmod | Associative Modifier | Common in Romanian/Baltic languages; shows nouns modifying other nouns. | nmod:assmod (Samochód mojego ojca) |
| :poss | Possessive | Indicates ownership or a possessive relationship. | nmod:poss (Mój pies, kapelusz Jana) |
| :relcl | Relative Clause | Identifies a clause that modifies a noun phrase. | acl:relcl (Książka, którą przeczytałem) |
| :tmod | Temporal Modifier | A modifier specifically describing time or duration. | nmod:tmod (Wyjeżdżam we wtorek) |
| :prt | Particle | Used for phrasal verb particles. | compound:prt (Poddaj się, wyłącz) |
| :rcomp | Relative Complement | Used for complements of relative clauses (common in Dutch). | advcl:rcomp (Mężczyzna, który wyszedł) |
| :flat | Flat Modifier | Used for multi-word expressions that don't have a clear internal head. | flat:name (Prezydent Obama) |
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.
| Label | Meaning | Example |
|---|---|---|
| 🌍 GPE | Geopolitical entity (countries, cities, states) | Polska, Warszawa, Francja, Kalifornia |
| 🏔️ LOC | Non-political location (mountains, rivers) | Ocean Spokojny, Mount Everest, Alpy |
| 🏢 FAC | Facility (buildings, airports, highways) | Most Golden Gate, Lotnisko Chopina, Burdż Chalifa |
| 👤 PERSON | People (real or fictional) | Elon Musk, Harry Potter, Alan Turing |
| 🚩 NORP | Nationalities, religious or political groups | Amerykanin, buddyzm, Demokraci, Japończyk |
| 🏢 ORG | Organizations (companies, institutions) | Google, Organizacja Narodów Zjednoczonych, Apple, FIFA |
| 📅 DATE | Absolute or relative dates | 4 lipca, 2026, wczoraj, w przyszłym tygodniu |
| ⌚ TIME | Times smaller than a day | 9:30 rano, zachód słońca, dziesięć minut |
| 🎊 EVENT | Named events (wars, festivals) | Druga wojna światowa, Coachella, Igrzyska Olimpijskie |
| 💰 MONEY | Monetary values, including unit | 100$, 5 milionów Euro, 50£ |
| ‱ PERCENT | Percentage, including "%" | 20%, osiemdziesiąt procent, 0,5% |
| ⚖️ QUANTITY | Measurements (weight, distance) | 5km, 50kg, 30 metrów kwadratowych |
| 🔢 ORDINAL | "First", "second", etc. | pierwszy, 2., dziewiąty |
| 🔢 CARDINAL | Numbers not classified elsewhere | 10, tysiąc, trzy |
| 📦 PRODUCT | Objects, vehicles, foods, etc. (not services) | iPhone, Tesla Model S, Coca-Cola |
| 🎨 WORK_OF_ART | Titles of books, songs, etc. | Mona Lisa, Bohemian Rhapsody, Hamlet |
| 📜 LAW | Named legal documents | Konstytucja, Traktat wersalski |
| 🗣️ LANGUAGE | Named languages | polski, Python, mandaryński |
Jeśli przetworzymy frazę „Google ma siedzibę w Kalifornii” (Google is based in California), warstwy analizy wyglądają następująco:
Lemat (Lemma): "Google", "mieć", "siedziba", "w", "Kalifornia"
UPOS: "PROPN(Nazwa własna)", "VERB(Czasownik)", "NOUN(Rzeczownik)", "ADP(Przyimek)", "PROPN(Nazwa własna)"
XPOS (NKJP): "subst:sg:nom:m3", "fin:sg:ter:imperf", "subst:sg:acc:f", "prep:loc:nwok", "subst:sg:loc:f"
DEP: „Google” to nsubj (podmiot nominalny) czasownika „ma”, który stanowi Root (główny czasownik w zdaniu). „siedzibę” to obj (dopełnienie bliższe), a „Kalifornii” to obl (okolicznik) połączony przyimkiem „w”.
NER: „Google” to 🏢 ORG (Organizacja), „Kalifornia” to 🌍 GPE (Jednostka geopolityczna).
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
|
|
|
© Stars21 - All Rights Reserved
|
|||||