{"id":1167524,"date":"2025-01-15T15:45:57","date_gmt":"2025-01-15T07:45:57","guid":{"rendered":"https:\/\/docs.pingcode.com\/ask\/ask-ask\/1167524.html"},"modified":"2025-01-15T15:45:59","modified_gmt":"2025-01-15T07:45:59","slug":"python%e5%a6%82%e4%bd%95%e5%af%bc%e5%85%a5%e8%af%ad%e6%96%99%e5%ba%93","status":"publish","type":"post","link":"https:\/\/docs.pingcode.com\/ask\/1167524.html","title":{"rendered":"python\u5982\u4f55\u5bfc\u5165\u8bed\u6599\u5e93"},"content":{"rendered":"

\"python\u5982\u4f55\u5bfc\u5165\u8bed\u6599\u5e93\"<\/p>\n

\u8981\u5728Python\u4e2d\u5bfc\u5165\u8bed\u6599\u5e93\uff0c\u4f60\u53ef\u4ee5\u4f7f\u7528NLTK\u3001spaCy\u3001Gensim<\/strong>\u7b49\u81ea\u7136\u8bed\u8a00\u5904\u7406\u5e93\u3002\u8fd9\u4e9b\u5e93\u63d0\u4f9b\u4e86\u5e7f\u6cdb\u7684\u8bed\u6599\u5e93\u548c\u6587\u672c\u5904\u7406\u5de5\u5177\uff0c\u80fd\u591f\u5e2e\u52a9\u4f60\u8fdb\u884c\u6587\u672c\u5206\u6790\u3001\u81ea\u7136\u8bed\u8a00\u5904\u7406\u548c\u673a\u5668\u5b66\u4e60<\/a>\u4efb\u52a1\u3002\u672c\u6587\u5c06\u8be6\u7ec6\u4ecb\u7ecd\u5982\u4f55\u4f7f\u7528\u8fd9\u4e09\u4e2a\u5e93\u5bfc\u5165\u548c\u4f7f\u7528\u8bed\u6599\u5e93\uff0c\u5e76\u63d0\u4f9b\u4e00\u4e9b\u793a\u4f8b\u4ee3\u7801\u6765\u5c55\u793a\u5b83\u4eec\u7684\u529f\u80fd\u3002<\/p>\n<\/p>\n

\u4e00\u3001NLTK\uff08Natural Language Toolkit\uff09<\/h3>\n<\/p>\n

NLTK\u662fPython\u4e2d\u6700\u6d41\u884c\u7684\u81ea\u7136\u8bed\u8a00\u5904\u7406\u5e93\u4e4b\u4e00\uff0c\u63d0\u4f9b\u4e86\u4e30\u5bcc\u7684\u8bed\u6599\u5e93\u548c\u6587\u672c\u5904\u7406\u5de5\u5177\u3002<\/p>\n<\/p>\n

1\u3001\u5b89\u88c5NLTK<\/h4>\n<\/p>\n

\u5728\u4f7f\u7528NLTK\u4e4b\u524d\uff0c\u4f60\u9700\u8981\u5148\u5b89\u88c5\u5b83\u3002\u4f60\u53ef\u4ee5\u4f7f\u7528\u4ee5\u4e0b\u547d\u4ee4\u5b89\u88c5NLTK\uff1a<\/p>\n<\/p>\n

pip install nltk<\/p>\n
<\/code><\/pre>\n<\/p>\n
2\u3001\u4e0b\u8f7d\u548c\u5bfc\u5165\u8bed\u6599\u5e93<\/h4>\n<\/p>\n
\u5b89\u88c5NLTK\u4e4b\u540e\uff0c\u4f60\u9700\u8981\u4e0b\u8f7d\u6240\u9700\u7684\u8bed\u6599\u5e93\u3002NLTK\u63d0\u4f9b\u4e86\u4e00\u4e2a\u65b9\u4fbf\u7684\u63a5\u53e3\u6765\u4e0b\u8f7d\u8bed\u6599\u5e93\u548c\u5176\u4ed6\u6570\u636e\u8d44\u6e90\u3002\u4ee5\u4e0b\u662f\u4e00\u4e2a\u793a\u4f8b\uff0c\u5c55\u793a\u5982\u4f55\u4e0b\u8f7d\u548c\u5bfc\u5165\u5e38\u7528\u7684\u8bed\u6599\u5e93\uff1a<\/p>\n<\/p>\n
import nltk<\/p>\n
nltk.download('all')  # \u4e0b\u8f7d\u6240\u6709NLTK\u6570\u636e\u96c6<\/p>\n
<\/code><\/pre>\n<\/p>\n
\u4f60\u53ef\u4ee5\u9009\u62e9\u4e0b\u8f7d\u7279\u5b9a\u7684\u8bed\u6599\u5e93\uff0c\u4f8b\u5982\uff1a<\/p>\n<\/p>\n
nltk.download('punkt')  # \u5206\u8bcd\u5de5\u5177<\/p>\n
nltk.download('wordnet')  # WordNet\u8bcd\u5178<\/p>\n
nltk.download('stopwords')  # \u505c\u7528\u8bcd\u8868<\/p>\n
<\/code><\/pre>\n<\/p>\n
3\u3001\u4f7f\u7528\u8bed\u6599\u5e93<\/h4>\n<\/p>\n
\u4e0b\u8f7d\u8bed\u6599\u5e93\u540e\uff0c\u4f60\u53ef\u4ee5\u4f7f\u7528NLTK\u63d0\u4f9b\u7684\u63a5\u53e3\u6765\u52a0\u8f7d\u548c\u5904\u7406\u8fd9\u4e9b\u8bed\u6599\u5e93\u3002\u4ee5\u4e0b\u662f\u4e00\u4e9b\u793a\u4f8b\u4ee3\u7801\uff1a<\/p>\n<\/p>\n
from nltk.corpus import gutenberg<\/p>\n
from nltk.corpus import stopwords<\/p>\n
from nltk.tokenize import word_tokenize<\/p>\n
\u52a0\u8f7d\u53e4\u767b\u5821\u8bed\u6599\u5e93\u4e2d\u7684\u67d0\u672c\u4e66<\/strong><\/h2>\n
text = gutenberg.raw('austen-emma.txt')<\/p>\n
print(text[:1000])  # \u6253\u5370\u524d1000\u4e2a\u5b57\u7b26<\/p>\n
\u5206\u8bcd<\/strong><\/h2>\n
tokens = word_tokenize(text)<\/p>\n
print(tokens[:50])  # \u6253\u5370\u524d50\u4e2a\u8bcd<\/p>\n
\u53bb\u9664\u505c\u7528\u8bcd<\/strong><\/h2>\n
stop_words = set(stopwords.words('english'))<\/p>\n
filtered_tokens = [word for word in tokens if word.lower() not in stop_words]<\/p>\n
print(filtered_tokens[:50])  # \u6253\u5370\u524d50\u4e2a\u8fc7\u6ee4\u540e\u7684\u8bcd<\/p>\n
<\/code><\/pre>\n<\/p>\n
\u4e8c\u3001spaCy<\/h3>\n<\/p>\n
spaCy\u662f\u53e6\u4e00\u4e2a\u5f3a\u5927\u7684\u81ea\u7136\u8bed\u8a00\u5904\u7406\u5e93\uff0c\u63d0\u4f9b\u4e86\u9ad8\u6548\u7684\u6587\u672c\u5904\u7406\u5de5\u5177\u548c\u9884\u8bad\u7ec3\u6a21\u578b\u3002<\/p>\n<\/p>\n
1\u3001\u5b89\u88c5spaCy<\/h4>\n<\/p>\n
\u5728\u4f7f\u7528spaCy\u4e4b\u524d\uff0c\u4f60\u9700\u8981\u5148\u5b89\u88c5\u5b83\u3002\u4f60\u53ef\u4ee5\u4f7f\u7528\u4ee5\u4e0b\u547d\u4ee4\u5b89\u88c5spaCy\uff1a<\/p>\n<\/p>\n
pip install spacy<\/p>\n
<\/code><\/pre>\n<\/p>\n
2\u3001\u4e0b\u8f7d\u548c\u5bfc\u5165\u8bed\u8a00\u6a21\u578b<\/h4>\n<\/p>\n
\u5b89\u88c5spaCy\u4e4b\u540e\uff0c\u4f60\u9700\u8981\u4e0b\u8f7d\u6240\u9700\u7684\u8bed\u8a00\u6a21\u578b\u3002spaCy\u63d0\u4f9b\u4e86\u591a\u4e2a\u9884\u8bad\u7ec3\u7684\u8bed\u8a00\u6a21\u578b\u3002\u4ee5\u4e0b\u662f\u4e00\u4e2a\u793a\u4f8b\uff0c\u5c55\u793a\u5982\u4f55\u4e0b\u8f7d\u548c\u5bfc\u5165\u8bed\u8a00\u6a21\u578b\uff1a<\/p>\n<\/p>\n
python -m spacy download en_core_web_sm  # \u4e0b\u8f7d\u5c0f\u578b\u7684\u82f1\u8bed\u8bed\u8a00\u6a21\u578b<\/p>\n
<\/code><\/pre>\n<\/p>\n
3\u3001\u4f7f\u7528\u8bed\u8a00\u6a21\u578b<\/h4>\n<\/p>\n
\u4e0b\u8f7d\u8bed\u8a00\u6a21\u578b\u540e\uff0c\u4f60\u53ef\u4ee5\u4f7f\u7528spaCy\u63d0\u4f9b\u7684\u63a5\u53e3\u6765\u52a0\u8f7d\u548c\u5904\u7406\u6587\u672c\u3002\u4ee5\u4e0b\u662f\u4e00\u4e9b\u793a\u4f8b\u4ee3\u7801\uff1a<\/p>\n<\/p>\n
import spacy<\/p>\n
\u52a0\u8f7d\u5c0f\u578b\u7684\u82f1\u8bed\u8bed\u8a00\u6a21\u578b<\/strong><\/h2>\n
nlp = spacy.load('en_core_web_sm')<\/p>\n
\u5904\u7406\u6587\u672c<\/strong><\/h2>\n
text = "Apple is looking at buying U.K. startup for $1 billion."<\/p>\n
doc = nlp(text)<\/p>\n
\u6253\u5370\u5206\u8bcd\u7ed3\u679c<\/strong><\/h2>\n
for token in doc:<\/p>\n
    print(token.text, token.lemma_, token.pos_, token.tag_, token.dep_, token.shape_, token.is_alpha, token.is_stop)<\/p>\n
\u6253\u5370\u547d\u540d\u5b9e\u4f53\u8bc6\u522b\u7ed3\u679c<\/strong><\/h2>\n
for ent in doc.ents:<\/p>\n
    print(ent.text, ent.start_char, ent.end_char, ent.label_)<\/p>\n
<\/code><\/pre>\n<\/p>\n
\u4e09\u3001Gensim<\/h3>\n<\/p>\n
Gensim\u662f\u4e00\u4e2a\u4e13\u6ce8\u4e8e\u4e3b\u9898\u5efa\u6a21\u548c\u6587\u6863\u76f8\u4f3c\u6027\u8ba1\u7b97\u7684\u81ea\u7136\u8bed\u8a00\u5904\u7406\u5e93\uff0c\u63d0\u4f9b\u4e86\u9ad8\u6548\u7684\u6587\u672c\u5904\u7406\u5de5\u5177\u548c\u9884\u8bad\u7ec3\u6a21\u578b\u3002<\/p>\n<\/p>\n
1\u3001\u5b89\u88c5Gensim<\/h4>\n<\/p>\n
\u5728\u4f7f\u7528Gensim\u4e4b\u524d\uff0c\u4f60\u9700\u8981\u5148\u5b89\u88c5\u5b83\u3002\u4f60\u53ef\u4ee5\u4f7f\u7528\u4ee5\u4e0b\u547d\u4ee4\u5b89\u88c5Gensim\uff1a<\/p>\n<\/p>\n
pip install gensim<\/p>\n
<\/code><\/pre>\n<\/p>\n
2\u3001\u5bfc\u5165\u548c\u4f7f\u7528\u9884\u8bad\u7ec3\u6a21\u578b<\/h4>\n<\/p>\n
Gensim\u63d0\u4f9b\u4e86\u591a\u4e2a\u9884\u8bad\u7ec3\u7684\u6a21\u578b\uff0c\u4f8b\u5982Word2Vec\u3001FastText\u548cDoc2Vec\u3002\u4ee5\u4e0b\u662f\u4e00\u4e9b\u793a\u4f8b\u4ee3\u7801\uff0c\u5c55\u793a\u5982\u4f55\u52a0\u8f7d\u548c\u4f7f\u7528\u8fd9\u4e9b\u6a21\u578b\uff1a<\/p>\n<\/p>\n
from gensim.models import Word2Vec<\/p>\n
\u52a0\u8f7d\u9884\u8bad\u7ec3\u7684Word2Vec\u6a21\u578b<\/strong><\/h2>\n
model = Word2Vec.load("path\/to\/your\/model")<\/p>\n
\u83b7\u53d6\u67d0\u4e2a\u8bcd\u7684\u8bcd\u5411\u91cf<\/strong><\/h2>\n
vector = model.wv['word']<\/p>\n
print(vector)<\/p>\n
\u627e\u5230\u4e0e\u67d0\u4e2a\u8bcd\u6700\u76f8\u4f3c\u7684\u8bcd<\/strong><\/h2>\n
similar_words = model.wv.most_similar('word', topn=10)<\/p>\n
print(similar_words)<\/p>\n
<\/code><\/pre>\n<\/p>\n
3\u3001\u4f7f\u7528Gensim\u5185\u7f6e\u7684\u8bed\u6599\u5e93<\/h4>\n<\/p>\n
Gensim\u8fd8\u63d0\u4f9b\u4e86\u4e00\u4e9b\u5185\u7f6e\u7684\u8bed\u6599\u5e93\uff0c\u4f60\u53ef\u4ee5\u4f7f\u7528\u8fd9\u4e9b\u8bed\u6599\u5e93\u8fdb\u884c\u8bad\u7ec3\u548c\u6d4b\u8bd5\u3002\u4ee5\u4e0b\u662f\u4e00\u4e2a\u793a\u4f8b\uff0c\u5c55\u793a\u5982\u4f55\u4f7f\u7528Gensim\u5185\u7f6e\u7684\u6587\u672c8\u8bed\u6599\u5e93\uff1a<\/p>\n<\/p>\n
import gensim<\/p>\n
from gensim.models import Word2Vec<\/p>\n
from gensim.test.utils import common_texts<\/p>\n
\u4f7f\u7528\u5185\u7f6e\u7684\u6587\u672c8\u8bed\u6599\u5e93\u8bad\u7ec3Word2Vec\u6a21\u578b<\/strong><\/h2>\n
model = Word2Vec(sentences=common_texts, vector_size=100, window=5, min_count=1, workers=4)<\/p>\n
\u4fdd\u5b58\u6a21\u578b<\/strong><\/h2>\n
model.save("word2vec.model")<\/p>\n
\u52a0\u8f7d\u6a21\u578b<\/strong><\/h2>\n
model = Word2Vec.load("word2vec.model")<\/p>\n
\u83b7\u53d6\u67d0\u4e2a\u8bcd\u7684\u8bcd\u5411\u91cf<\/strong><\/h2>\n
vector = model.wv['computer']<\/p>\n
print(vector)<\/p>\n
\u627e\u5230\u4e0e\u67d0\u4e2a\u8bcd\u6700\u76f8\u4f3c\u7684\u8bcd<\/strong><\/h2>\n
similar_words = model.wv.most_similar('computer', topn=10)<\/p>\n
print(similar_words)<\/p>\n
<\/code><\/pre>\n<\/p>\n
\u901a\u8fc7\u4e0a\u8ff0\u6b65\u9aa4\uff0c\u4f60\u53ef\u4ee5\u5728Python\u4e2d\u8f7b\u677e\u5bfc\u5165\u548c\u4f7f\u7528\u8bed\u6599\u5e93\u8fdb\u884c\u81ea\u7136\u8bed\u8a00\u5904\u7406\u4efb\u52a1\u3002\u6839\u636e\u4f60\u7684\u9700\u6c42\u9009\u62e9\u5408\u9002\u7684\u5e93\u548c\u8bed\u6599\u5e93\uff0c\u53ef\u4ee5\u5e2e\u52a9\u4f60\u66f4\u9ad8\u6548\u5730\u8fdb\u884c\u6587\u672c\u5206\u6790\u548c\u673a\u5668\u5b66\u4e60\u3002<\/p>\n<\/p>\n
\u76f8\u5173\u95ee\u7b54FAQs\uff1a<\/strong><\/h2>\n
 \u5982\u4f55\u5728Python\u4e2d\u5bfc\u5165\u548c\u4f7f\u7528\u8bed\u6599\u5e93\uff1f<\/strong>
\u5728Python\u4e2d\uff0c\u5bfc\u5165\u8bed\u6599\u5e93\u901a\u5e38\u53ef\u4ee5\u901a\u8fc7\u4f7f\u7528\u7279\u5b9a\u7684\u5e93\u6765\u5b9e\u73b0\u3002\u4f8b\u5982\uff0cNLTK\uff08\u81ea\u7136\u8bed\u8a00\u5de5\u5177\u5305\uff09\u662f\u4e00\u4e2a\u5e38\u7528\u7684\u5e93\uff0c\u5b83\u63d0\u4f9b\u4e86\u591a\u79cd\u8bed\u6599\u5e93\u548c\u5de5\u5177\u3002\u9996\u5148\uff0c\u60a8\u9700\u8981\u5b89\u88c5NLTK\u5e93\uff0c\u7136\u540e\u4f7f\u7528nltk.download()<\/code>\u547d\u4ee4\u4e0b\u8f7d\u6240\u9700\u7684\u8bed\u6599\u5e93\u3002\u5b8c\u6210\u540e\uff0c\u53ef\u4ee5\u901a\u8fc7nltk.corpus<\/code>\u6a21\u5757\u8f7b\u677e\u8bbf\u95ee\u548c\u4f7f\u7528\u8fd9\u4e9b\u8bed\u6599\u5e93\u3002<\/p>\n
\u6211\u53ef\u4ee5\u5728\u54ea\u91cc\u627e\u5230\u9002\u5408\u6211\u7684\u9879\u76ee\u7684\u8bed\u6599\u5e93\uff1f<\/strong>
\u6709\u8bb8\u591a\u5728\u7ebf\u8d44\u6e90\u63d0\u4f9b\u4e0d\u540c\u9886\u57df\u7684\u8bed\u6599\u5e93\u3002\u5e38\u89c1\u7684\u9009\u62e9\u5305\u62ecNLTK\u3001spaCy\u3001Gensim\u4ee5\u53caKaggle\u7b49\u5e73\u53f0\u3002\u60a8\u53ef\u4ee5\u6839\u636e\u9879\u76ee\u9700\u6c42\u9009\u62e9\u5408\u9002\u7684\u8bed\u6599\u5e93\uff0c\u6bd4\u5982\u6587\u672c\u5206\u7c7b\u3001\u60c5\u611f\u5206\u6790\u7b49\u3002\u540c\u65f6\uff0c\u8bb8\u591a\u5b66\u672f\u673a\u6784\u548c\u7814\u7a76\u4eba\u5458\u4e5f\u4f1a\u5171\u4eab\u4ed6\u4eec\u7684\u8bed\u6599\u5e93\uff0c\u60a8\u53ef\u4ee5\u901a\u8fc7Google Scholar\u6216\u76f8\u5173\u7684\u7814\u7a76\u8bba\u6587\u627e\u5230\u94fe\u63a5\u3002<\/p>\n
\u5bfc\u5165\u8bed\u6599\u5e93\u540e\uff0c\u6211\u8be5\u5982\u4f55\u5904\u7406\u548c\u5206\u6790\u6570\u636e\uff1f<\/strong>
\u5bfc\u5165\u8bed\u6599\u5e93\u540e\uff0c\u53ef\u4ee5\u4f7f\u7528\u591a\u79cd\u6570\u636e\u5904\u7406\u548c\u5206\u6790\u5de5\u5177\u3002Python\u7684pandas\u5e93\u975e\u5e38\u9002\u5408\u5904\u7406\u7ed3\u6784\u5316\u6570\u636e\uff0c\u800cNLTK\u548cspaCy\u5219\u63d0\u4f9b\u4e86\u4e00\u4e9b\u5f3a\u5927\u7684\u81ea\u7136\u8bed\u8a00\u5904\u7406\u529f\u80fd\uff0c\u5982\u8bcd\u6027\u6807\u6ce8\u3001\u547d\u540d\u5b9e\u4f53\u8bc6\u522b\u7b49\u3002\u60a8\u53ef\u4ee5\u5148\u5bf9\u6570\u636e\u8fdb\u884c\u6e05\u6d17\u548c\u9884\u5904\u7406\uff0c\u7136\u540e\u4f7f\u7528\u53ef\u89c6\u5316\u5de5\u5177\uff08\u5982Matplotlib\u548cSeaborn\uff09\u6765\u5206\u6790\u548c\u5c55\u793a\u7ed3\u679c\uff0c\u5e2e\u52a9\u60a8\u66f4\u597d\u5730\u7406\u89e3\u6570\u636e\u4e2d\u7684\u6a21\u5f0f\u548c\u8d8b\u52bf\u3002<\/p>\n","protected":false},"excerpt":{"rendered":"\u8981\u5728Python\u4e2d\u5bfc\u5165\u8bed\u6599\u5e93\uff0c\u4f60\u53ef\u4ee5\u4f7f\u7528NLTK\u3001spaCy\u3001Gensim\u7b49\u81ea\u7136\u8bed\u8a00\u5904\u7406\u5e93\u3002\u8fd9\u4e9b\u5e93\u63d0\u4f9b\u4e86\u5e7f\u6cdb\u7684 […]","protected":false},"author":3,"featured_media":1167527,"comment_status":"closed","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"footnotes":""},"categories":[37],"tags":[],"acf":[],"_links":{"self":[{"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/posts\/1167524"}],"collection":[{"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/users\/3"}],"replies":[{"embeddable":true,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/comments?post=1167524"}],"version-history":[{"count":"1","href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/posts\/1167524\/revisions"}],"predecessor-version":[{"id":1167528,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/posts\/1167524\/revisions\/1167528"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/media\/1167527"}],"wp:attachment":[{"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/media?parent=1167524"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/categories?post=1167524"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/tags?post=1167524"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}