ADV

home zh/pos edit page issue tracker

This page pertains to UD version 2.

`ADV`: adverb

Definition

Adverbs (副詞 / fùcí) typically modify verbs, adjectives, and other adverbs for such categories as time, manner, degree, frequency, or negation.

你慢慢/ADV 説/VERB / nǐ mànman shuō “Speak slowly.”
她非常/ADV 快樂/ADJ / tā fēicháng kuàilè “She is extremely happy.”
他們很/ADV 早/ADV 起來/VERB / tāmen hěn zǎo qǐlái “They get up very early.”

Some adverbs also modify clauses with conjunctive and discursive functions.

他卻沒有跟我説 / tā què méiyǒu gēn wǒ shuō “He didn’t tell me, however.”
今天當然沒人呀／ jīntiān dāngrán méi rén ya “Of course nobody’s here today.”

A small number of adverbs may also modify numerals and determiners, or nouns and pronouns.

差不多 五千 / chàbùduō wǔqiān “almost five thousand”
連他都不去了 / lián tā dōu bù qù le “Even he’s not going.”

There is a closed subclass of pronominal adverbs that refer to circumstances in context, rather than naming them directly; similarly to pronouns, these can be categorized as interrogative, demonstrative, etc. These should be treated as adverbs when modifying a predicate, but otherwise some of them can function as a nominal in the syntax, in which case they should be tagged PRON.

你 為什麼 不來 / nǐ wèishéme bù lái “Why aren’t you coming?”
我這樣做的 / wǒ shì zhèyàng zuò de “I do it like this.”

Note that although some adverbs express temporal information, many common time expressions (e.g., 今天 / jīntiān ‘today’, 去年 / qùnián “last year”, 晚上 / wǎnshàng “night”) are actually nouns and should be tagged NOUN.

Examples

Manner adverbs
- 慢慢 / mànman “slowly”, 互相／ hùxiāng “mutually”
Temporal, aspectual, and modal adverbs
- 即將 / jíjiāng “about to”, 將 / jiāng “will”, (正)在 / (zhèng)zài “in progress”, 寧可 / nìngkě “would rather”, 可能 / kěnéng “possibly” (NB: also ADJ “possible” and NOUN “possibility”)
Conjunctive adverbs
- 所以 / suǒyǐ “therefore”, 卻 / què “however”, 例如 / lìrú “for example”, 然後 / ránhòu “afterwards”, 那 / nà “then; in that case”, 就 / jiù “then; as a result”
Frequency and duration adverbs
- 經常 / jīngcháng “often”, 一下 / yīxià “once; for a little while”, 多次 / duōcì “many times”
Negation adverbs
- 不 / bù “not”, 未 / wèi “not yet”, 永不 / yóngbù “never”
Numeral-modifying adverbs
- 大概 / dàgài “approximately”, 差不多 / chàbùduō “almost”, 不足 / bùzú “less than”, 至少 / zhìshǎo “at least”, 最多 / zuìduō “at most”
Noun-modifying adverbs
- 連 / lián “even”, 又 / yòu “another”
Pronominal adverbs
- 怎樣 / zěnyàng “how”, 這樣 / zhèyàng “this way, like this, thus”, 那麼 / nàme “so”, 為什麼 / wèishéme “why”
Other
- 都 / dōu “also; all”, 也 / yě “also”, 才 / cái “then and only then”, 越來越 yüèláiyüè “more and more”, 當然 / dāngrán “of course”, 難道 / nándào “could/must it be that…?”
- “for example”, “such as”: 比如 / bǐrú, 比如說 / bǐrúshuō, 例如 / lìrú, 譬如 / pìrú,
- “like”: 像 / xiàng (note this can also function as a main verb; an example of the adverbial usage is 像他昨天說… / xiàng tā zuótiān shuō “Like he said yesterday…”)

Treebank Statistics (UD_Chinese)

There are 489 ADV lemmas (2%), 489 ADV types (2%) and 5200 ADV tokens (5%). Out of 15 observed tags, the rank of ADV is: 8 in number of lemmas, 8 in number of types and 7 in number of tokens.

The 10 most frequent ADV lemmas: 也、並、而、但、不、所、都、曾、則、就

The 10 most frequent ADV types: 也、並、而、但、不、所、都、曾、則、就

The 10 most frequent ambiguous lemmas: 也 (ADV 290, CCONJ 6), 並 (ADV 283, CCONJ 14), 而 (ADV 283, CCONJ 11), 不 (ADV 180, PART 3), 所 (ADV 138, NOUN 22, PART 21), 都 (ADV 133, PART 1, NOUN 1), 曾 (ADV 105, PROPN 8), 則 (ADV 102, NOUN 3), 就 (ADV 96, ADP 4), 已 (ADV 95, VERB 1)

The 10 most frequent ambiguous types: 也 (ADV 290, CCONJ 6), 並 (ADV 283, CCONJ 14), 而 (ADV 283, CCONJ 11), 不 (ADV 180, PART 3), 所 (ADV 138, NOUN 22, PART 21), 都 (ADV 133, PART 1, NOUN 1), 曾 (ADV 105, PROPN 8), 則 (ADV 102, NOUN 3), 就 (ADV 96, ADP 4), 已 (ADV 95, VERB 1)

也
- ADV 290: 同樣 , 施力的大小不同 , 引起的加速度不同 , 最終的結果也不一樣 , 亦可以從向量的加成性來看 .
- CCONJ 6: 合作的形式 , 既有版權授權 , 也有中外雙方共同開發選題合作出版 .
並
- ADV 283: 史萊姆是由鳥山明為《勇者斗惡龍》設計的 , 並成為了系列官方的吉祥物 .
- CCONJ 14: 交換的不斷發展和擴大 , 使商品生產出現並發展 , 又反過來促進了交換的進一步發展 .
而
- ADV 283: 數百萬的巧克力棒被撤下櫃檯 , 而瑪氏則中斷了生產 , 公司損失達四百五十萬美元 .
- CCONJ 11: 卡萊爾是擁有簡潔而歷史意義重大的城市 , 該地有一座城堡、博物館、大教堂以及不完整的城牆 .
不
- ADV 180: 同樣 , 施力的大小不同 , 引起的加速度不同 , 最終的結果也不一樣 , 亦可以從向量的加成性來看 .
- PART 3: 面對不可能和疑問 , 理性的人就只有放棄或否認 , 並永遠為失去的東西而悲傷 .
所
- ADV 138: 包盈盈所在的第三區 , 有衛冕冠軍、 6 號種子瑪麗埃爾 · 扎格尼斯 .
- NOUN 22: 只是在紐約州 , 就有 500多所學校將匹克球設為課程 .
- PART 21: 於是羅馬的異端裁判所成立 , 由羅馬教廷控制 .
都
- ADV 133: 但是迪士尼的公主們不會都太侷限於一個範圍之內了嗎 ?
- PART 1: 新宿御苑是日本東京都橫跨新宿區與澀谷區的庭園 , 面積 58.3 公頃 .
- NOUN 1: 比較有代表性的有北京大學中文系袁毓林先生 , 比如他對漢語「都」的研究 .
曾
- ADV 105: 理學家朱熹任漳州知府時 , 也曾禁止當地戲曲演出 .
- PROPN 8: 曾成傑案是湘西非法集資系列案件中的一件 .
則
- ADV 102: 數百萬的巧克力棒被撤下櫃檯 , 而瑪氏則中斷了生產 , 公司損失達四百五十萬美元 .
- NOUN 3: 這則故事的原型 , 很可能是出自《舊五代史 · 武皇紀上》的記載 .
就
- ADV 96: 只是在紐約州 , 就有 500多所學校將匹克球設為課程 .
- ADP 4: 但至少就當時而言 , 這項改革還是鞏固了殖民地的基礎 .
已
- ADV 95: 努克的經濟以漁業為主 , 1960 年代 , 鱈魚業相當發達 , 現已沒落 , 目前主要出產螃蟹和大比目魚 .
- VERB 1: 雖然東亞銀行在揭發做假帳事件後已採取措施解決 , 但仍有人擔心東亞的財政狀況 , 甚至製造謠言 .

Morphology

The form / lemma ratio of ADV is 1.000000 (the average of all parts of speech is 1.000284).

The 1st highest number of forms (1) was observed with the lemma “,”: ,.

The 2nd highest number of forms (1) was observed with the lemma “一併”: 一併.

The 3rd highest number of forms (1) was observed with the lemma “一共”: 一共.

ADV occurs with 1 features: zh-feat/Polarity (218; 4% instances)

ADV occurs with 1 feature-value pairs: Polarity=Neg

ADV occurs with 2 feature combinations. The most frequent feature combination is _ (4982 tokens). Examples: 也、並、而、但、所、都、曾、則、就、已

Relations

ADV nodes are attached to their parents using 12 different relations: zh-dep/advmod (2761; 53% instances), zh-dep/mark (2415; 46% instances), zh-dep/acl (5; 0% instances), zh-dep/amod (4; 0% instances), zh-dep/nmod:tmod (4; 0% instances), zh-dep/root (4; 0% instances), zh-dep/ccomp (2; 0% instances), zh-dep/advcl (1; 0% instances), zh-dep/conj (1; 0% instances), zh-dep/dep (1; 0% instances), zh-dep/punct (1; 0% instances), zh-dep/xcomp (1; 0% instances)

Parents of ADV nodes belong to 14 different parts of speech: VERB (4337; 83% instances), ADJ (473; 9% instances), NOUN (243; 5% instances), ADV (47; 1% instances), NUM (29; 1% instances), PART (26; 1% instances), AUX (20; 0% instances), ADP (8; 0% instances), PROPN (7; 0% instances), ROOT (4; 0% instances), DET (3; 0% instances), PRON (1; 0% instances), SYM (1; 0% instances), X (1; 0% instances)

5134 (99%) ADV nodes are leaves.

45 (1%) ADV nodes have one child.

13 (0%) ADV nodes have two children.

8 (0%) ADV nodes have three or more children.

The highest child degree of a ADV node is 7.

Children of ADV nodes are attached using 14 different relations: zh-dep/advmod (43; 41% instances), zh-dep/case (12; 11% instances), zh-dep/punct (10; 9% instances), zh-dep/xcomp (10; 9% instances), zh-dep/ccomp (7; 7% instances), zh-dep/mark (5; 5% instances), zh-dep/obj (5; 5% instances), zh-dep/nsubj (4; 4% instances), zh-dep/advcl (2; 2% instances), zh-dep/cc (2; 2% instances), zh-dep/nmod (2; 2% instances), zh-dep/nmod:tmod (2; 2% instances), zh-dep/csubj (1; 1% instances), zh-dep/dep (1; 1% instances)

Children of ADV nodes belong to 10 different parts of speech: ADV (47; 44% instances), VERB (17; 16% instances), ADP (12; 11% instances), NOUN (11; 10% instances), PUNCT (10; 9% instances), ADJ (3; 3% instances), CCONJ (2; 2% instances), PROPN (2; 2% instances), PART (1; 1% instances), PRON (1; 1% instances)