stanzaで言語分析 2

stanzaを使ってみる 2

---

以前の投稿でstanzaを使ってみたことをまとめた．

今回はその続きで品詞のタグ付けと単語情報の抽出をstanzaで行ってみたい．

コードは下記．

import stanza

stanza.download('en')
nlp = stanza.Pipeline('en')  # デフォルトではtokenize, mwt, pos, lemma, depparseが有効

text = "Artificial intelligence is transforming the world."

doc = nlp(text)

for sentence in doc.sentences:
    for word in sentence.words:
        print(f"Word: {word.text}")
        print(f"  Lemma: {word.lemma}")  # 原形
        print(f"  POS: {word.upos}")     # Universal POS tag (名詞、動詞など)
        print(f"  XPOS: {word.xpos}")    # 言語特有の詳細な品詞

Output例

Word: Artificial

Lemma: artificial

POS: ADJ

XPOS: JJ

Word: intelligence

Lemma: intelligence

POS: NOUN

XPOS: NN

Word: is

Lemma: be

POS: AUX

XPOS: VBZ

Word: transforming

Lemma: transform

POS: VERB

XPOS: VBG

Word: the

Lemma: the

POS: DET

XPOS: DT

Word: world

Lemma: world

POS: NOUN

XPOS: NN

Word: .

Lemma: .

POS: PUNCT

XPOS: .