Python 中的 NLP 特征工程
Rounak Banik
Data Scientist
'I am happy'
'I am joyous'
'I am sad'
King-Queen → Man-WomanFrance-Paris → Russia-Moscowimport spacy
# 加载模型并创建 Doc 对象
nlp = spacy.load('en_core_web_lg')
doc = nlp('I am happy')
# 为每个 token 生成词向量
for token in doc:
print(token.vector)
[-1.0747459e+00 4.8677087e-02 5.6630421e+00 1.6680446e+00
-1.3194644e+00 -1.5142369e+00 1.1940931e+00 -3.0168812e+00
...
doc = nlp("happy joyous sad")
for token1 in doc:
for token2 in doc:
print(token1.text, token2.text, token1.similarity(token2))
happy happy 1.0
happy joyous 0.63244456
happy sad 0.37338886
joyous happy 0.63244456
joyous joyous 1.0
joyous sad 0.5340932
...
# 生成 doc 对象
sent1 = nlp("I am happy")
sent2 = nlp("I am sad")
sent3 = nlp("I am joyous")
# 计算 sent1 与 sent2 的相似度
sent1.similarity(sent2)
0.9273363837282105
# 计算 sent1 与 sent3 的相似度
sent1.similarity(sent3)
0.9403554938594568
Python 中的 NLP 特征工程