Skip to content

Linguistic Reference

All transcript data in TCSE is annotated using spaCy 3.8 (en_core_web_lg). This page provides a comprehensive reference for all linguistic annotation categories used in TCSE's Advanced Search.


Universal POS Tags

Coarse-grained part-of-speech categories based on Universal Dependencies. Used in Advanced Search with {pos} notation (e.g., help{verb}, [be]{aux}).

Shorthand aliases are available for frequently used tags — for example, {v} can be used instead of {verb}.

Tag Part of Speech Search Aliases Type Examples
ADJ Adjective {adj} {a}, {j}, {jj} Open big, old, green, first
ADV Adverb {adv} {r}, {rb} Open very, well, exactly, tomorrow
INTJ Interjection {intj} Open hello, ouch, bravo
NOUN Noun {noun} {n}, {nn} Open people, time, world, way
PROPN Proper noun {propn} Open Mary, John, London, NATO
VERB Verb {verb} {v}, {vb} Open run, eat, go, think, help
ADP Adposition {adp} {p} Closed in, to, during, of, with
AUX Auxiliary {aux} Closed has, is, will, should, must
CCONJ Coordinating conjunction {cconj} {c} Closed and, or, but
DET Determiner {det} {dt} Closed the, a, this, which, all
NUM Numeral {num} Closed one, two, three
PART Particle {part} Closed 's, not, to
PRON Pronoun {pron} {pr} Closed I, it, you, we, they
SCONJ Subordinating conjunction {sconj} Closed that, if, while, because
PUNCT Punctuation {punct} Other . , ; : ! ?
SYM Symbol {sym} Other $, %, §
X Other {x} Other foreign words, typos
SPACE Space {space} Other whitespace tokens

Fine-Grained Tags (Penn Treebank)

Detailed POS tags following the Penn Treebank tagset. Searchable with {@tag} notation (e.g., {@VBG} for gerunds/present participles).

Nouns

Tag Description Examples
NN Noun, singular or mass dog, music, education
NNS Noun, plural dogs, children, ideas
NNP Proper noun, singular London, Obama, Google
NNPS Proper noun, plural Americans, Alps

Verbs

Tag Description Examples
VB Verb, base form go, eat, run
VBD Verb, past tense went, ate, ran
VBG Verb, gerund / present participle going, eating, running
VBN Verb, past participle gone, eaten, taken
VBP Verb, non-3rd person singular present go, eat, run
VBZ Verb, 3rd person singular present goes, eats, runs
MD Modal can, could, may, might, will, would, shall, should, must

Adjectives

Tag Description Examples
JJ Adjective big, green, incomprehensible
JJR Adjective, comparative bigger, better, faster
JJS Adjective, superlative biggest, best, fastest
AFX Affix (hyphenated adjective part) e.g., "anti" in "anti-war"

Adverbs

Tag Description Examples
RB Adverb very, well, quickly, never
RBR Adverb, comparative faster, better, more
RBS Adverb, superlative fastest, best, most
WRB Wh-adverb where, when, why, how

Pronouns & Determiners

Tag Description Examples
PRP Personal pronoun I, you, he, she, it, we, they
PRP$ Possessive pronoun my, your, his, her, its, our, their
WP Wh-pronoun who, what, whom
WP$ Possessive wh-pronoun whose
WDT Wh-determiner which, that, what
DT Determiner the, a, an, this, that, these
PDT Predeterminer all, both, half
EX Existential there there (is/are)

Other

Tag Description Examples
IN Preposition or subordinating conjunction in, of, by, that, if, because
CC Coordinating conjunction and, or, but, nor, yet
CD Cardinal number one, 2, 1,000
TO to to (go), to (the store)
POS Possessive ending 's, '
RP Particle up, off, out, in (phrasal verbs)
UH Interjection oh, well, um, uh
FW Foreign word de, la, von
LS List item marker 1., 2., a., b.
SYM Symbol $, %, +, =
HYPH Hyphen -
NFP Superfluous punctuation ..., --
XX Unknown unanalyzable tokens
ADD Email or URL user@example.com
_SP Space whitespace

Dependency Labels

Syntactic dependency relations showing how words relate to each other in a sentence. Searchable with {@label} notation (e.g., {@nsubj} for nominal subjects).

Note

The {@...} notation searches both fine-grained tags and dependency labels simultaneously. For example, {@VBG} matches gerunds (tag) while {@nsubj} matches nominal subjects (dependency).

Core Arguments

Label Description Example
nsubj Nominal subject She runs.
nsubjpass Nominal subject (passive) It was built.
dobj Direct object I see you.
dative Dative (indirect object) Give me a book.
attr Attribute She is a teacher.
agent Agent (passive by-phrase) Built by engineers.
expl Expletive There is a problem.

Clausal Arguments

Label Description Example
csubj Clausal subject What she said is true.
csubjpass Clausal subject (passive) That he came was unexpected.
ccomp Clausal complement I think he left.
xcomp Open clausal complement I want to go.
acomp Adjectival complement She looks happy.
oprd Object predicate I consider him smart.

Modifiers

Label Description Example
amod Adjectival modifier a big house
advmod Adverbial modifier run quickly**
nummod Numeric modifier three cats
nmod Nominal modifier a cup of coffee**
npadvmod Noun phrase as adverbial modifier yesterday*, this way***
quantmod Quantifier modifier about 200**
appos Appositional modifier Sam, my brother**
acl Adjectival/relative clause the man who came**
relcl Relative clause modifier the book I read**
advcl Adverbial clause modifier If it rains, I stay.
neg Negation modifier I do not agree.
det Determiner the book
predet Predeterminer all the people
poss Possession modifier my book

Prepositional & Case

Label Description Example
prep Prepositional modifier go to school
pobj Object of preposition in the house**
pcomp Complement of preposition instead of going**
case Case marking John 's book

Coordination & Connectors

Label Description Example
conj Conjunct cats and dogs**
cc Coordinating conjunction cats and dogs
preconj Pre-correlative conjunction either A or B
mark Marker (subordinating conjunction) because it rained

Verbal

Label Description Example
aux Auxiliary I have eaten.
auxpass Passive auxiliary It was built.
compound Compound New* York, ice* cream
prt Particle (phrasal verb) give up, turn off**

Other

Label Description Example
ROOT Root of the sentence She runs fast.
punct Punctuation Hello, world.**
parataxis Parataxis (loosely joined clause) He said — I agree.
intj Interjection Well, I think...
dep Unclassified dependent (catch-all)
meta Meta modifier structural markup

Morphological Features

Grammatical properties of individual tokens. Searchable with {#feature} notation using partial matching (e.g., {#past} matches any token containing "Past" in its morphological annotation).

Feature Values Search example Description
VerbForm Fin, Ger, Inf, Part {#ger}, {#inf} Finite, gerund, infinitive, participle
Tense Past, Pres {#past}, {#pres} Past or present tense
Aspect Perf, Prog {#perf}, {#prog} Perfect or progressive aspect
Mood Ind {#ind} Indicative mood
VerbType Mod {#mod} Modal verb
Number Sing, Plur {#sing}, {#plur} Singular or plural
Person 1, 2, 3 {#person: 3} 1st, 2nd, or 3rd person
Case Acc, Nom {#nom}, {#acc} Nominative or accusative case
Gender Fem, Masc, Neut {#fem}, {#masc} Grammatical gender (pronouns)
Degree Pos, Cmp, Sup {#cmp}, {#sup} Positive, comparative, superlative
Definite Def, Ind {#def}, {#ind} Definite or indefinite article
PronType Art, Dem, Ind, Prs, Rel {#dem}, {#rel} Article, demonstrative, indefinite, personal, relative
Poss Yes {#poss} Possessive
Reflex Yes {#reflex} Reflexive pronoun
Polarity Neg {#neg} Negative polarity (not)
NumType Card, Mult, Ord {#ord} Cardinal, multiplicative, ordinal
Foreign Yes {#foreign} Foreign word
ConjType Cmp {#conjtype} Comparative conjunction (than)

Passive voice

spaCy's en_core_web_lg model does not annotate passive voice in the morphological features for English. Use dependency labels instead: {@auxpass} (passive auxiliary) or {@nsubjpass} (passive nominal subject).


Named Entity Types

Named entities recognized by spaCy's NER model. Searchable with %TYPE notation (e.g., %PERSON, say %ORG). See also Named Entity Search.

Type Description Examples
PERSON People, including fictional Obama, Einstein, Hamlet
ORG Organizations, companies, agencies Google, the UN, NASA
GPE Countries, cities, states France, New York, California
LOC Non-GPE locations: mountains, bodies of water the Alps, the Pacific, Antarctica
FAC Facilities: buildings, airports, highways the Eiffel Tower, JFK Airport
NORP Nationalities, religious/political groups American, Buddhist, Republican
DATE Absolute or relative dates/periods tomorrow, last year, 2024
TIME Times shorter than a day 3 o'clock, this morning
MONEY Monetary values $5, 2 million euros
PERCENT Percentages fifty percent, 10%
QUANTITY Measurements: weight, distance, etc. 5 kg, 2 miles, 100 degrees
CARDINAL Numerals not covered by other types five, hundred, 42
ORDINAL Ordinal numbers first, 42nd, third
EVENT Named events World War II, the Olympics
WORK_OF_ART Titles of works Hamlet, Mona Lisa
PRODUCT Products, vehicles, foods iPhone, Boeing 747
LAW Named documents of law the Magna Carta, GDPR
LANGUAGE Named languages English, Mandarin, Arabic

Advanced Search Syntax Summary

Quick reference for all search notation available in Advanced Search mode. For detailed usage and examples, see Advanced Search Query Syntax.

Syntax Description Example
word Surface form search help
[lemma] Lemma (base form) search [be] → am, is, are, was, were, been, being
['s] Literal surface match ['s] → 's (possessive/contraction)
word{pos} Surface + POS filter help{verb}, help{noun}
[lemma]{pos} Lemma + POS filter [help]{noun}
{pos} POS-only search {propn} (any proper noun)
{@tag_or_dep} Fine-grained tag or dependency {@VBG} (gerund), {@nsubj} (subject)
{#morph} Morphological feature {#past} (past tense), {#plur} (plural)
{-pos} Negative POS filter {-verb} (not a verb)
-word Negative word match -the
a\|b OR alternatives help\|assist
+prefix Prefix match +un → un..., under..., until...
{} or * Wildcard (one or more words) [make] {} {noun}
-_ Wildcard (exactly one word) to -_ surprise
_ Noun chunk placeholder [give] _ _
^ Start of segment ^ however
%TYPE Named entity search %PERSON, say %ORG