このページの2つのバージョン間の差分を表示します。
両方とも前のリビジョン前のリビジョン次のリビジョン | 前のリビジョン | ||
ライブラリ:ginza [2022/04/29 14:24] – ↷ 移動操作に合わせてリンクを書き換えました。 admin | ライブラリ:ginza [2023/10/03 18:03] (現在) – admin | ||
---|---|---|---|
行 3: | 行 3: | ||
オープンソース日本語自然言語処理[[: | オープンソース日本語自然言語処理[[: | ||
- | ==== Google Colab上でちょっと使ってみる | + | * [[ライブラリ: |
- | === 準備 === | + | |
- | + | ||
- | !pip install -U ginza # GiNZAのインストール | + | |
- | import pkg_resources, | + | |
- | imp.reload(pkg_resources) | + | |
- | import spacy # SpaCy | + | |
- | nlp = spacy.load(' | + | |
- | from spacy import displacy | + | |
- | + | ||
- | === 形態素解析 === | + | |
- | + | ||
- | doc = nlp(' | + | |
- | for sent in doc.sents: | + | |
- | for token in sent: | + | |
- | print(token.i, | + | |
- | + | ||
- | === 固有表現認識 === | + | |
- | + | ||
- | doc = nlp(' | + | |
- | displacy.render(doc, | + | |
- | + | ||
- | === 構文解析 === | + | |
- | + | ||
- | doc = nlp(' | + | |
- | displacy.render(doc, | + | |
- | + | ||
- | === 単語の類似度を計算 === | + | |
- | + | ||
- | tokens = nlp(' | + | |
- | for t1 in tokens: | + | |
- | for t2 in tokens: | + | |
- | if t1 == t2: | + | |
- | break | + | |
- | print(' | + | |
- | + | ||
- | === 文の類似度を計算 === | + | |
- | + | ||
- | doc1 = nlp(' | + | |
- | doc2 = nlp(' | + | |
- | doc3 = nlp(' | + | |
- | for d1 in (doc1, doc2, doc3): | + | |
- | for d2 in (doc1, doc2, doc3): | + | |
- | if d1 == d2: | + | |
- | break | + | |
- | print(' | + | |
=== 記事 === | === 記事 === | ||
+ | * 2022-12-08 | [[https:// | ||
+ | * 2022-09-16 | [[https:// | ||
+ | * 2022-09-16 | [[https:// | ||
+ | * 2022-08-09 | [[https:// | ||
+ | * 2022-07-31 | [[https:// | ||
+ | * 2022-07-22 | [[https:// | ||
+ | * 2022-07-11 | [[https:// | ||
+ | * 2022-05-15 | [[https:// | ||
+ | * 2022-03-29 | [[https:// | ||
* 2022-02-28 | [[https:// | * 2022-02-28 | [[https:// | ||
* 2022-02-27 | [[https:// | * 2022-02-27 | [[https:// | ||
行 64: | 行 27: | ||
* 2020-12-16 | [[https:// | * 2020-12-16 | [[https:// | ||
* 2020-10-10 | [[https:// | * 2020-10-10 | [[https:// | ||
- | * 2020-09-27 | [[https:// | + | * 2020-09-27 | [[https:// |
* 2020-07-07 | [[https:// | * 2020-07-07 | [[https:// | ||
* 2020-03-24 | [[https:// | * 2020-03-24 | [[https:// |