scikit cosine_similarity vs pairwise_distances

Question

Scikit-learn的sklearn.metrics.pairwise.cosine_similarity和sklearn.metrics.pairwise.pairwise_distances(..的区别是什么Metric=“余弦”)？

from sklearn.feature_extraction.text import TfidfVectorizer

documents = (
    "Macbook Pro 15' Silver Gray with Nvidia GPU",
    "Macbook GPU"    
)

tfidf_vectorizer = TfidfVectorizer()
tfidf_matrix = tfidf_vectorizer.fit_transform(documents)

from sklearn.metrics.pairwise import cosine_similarity
print(cosine_similarity(tfidf_matrix[0:1], tfidf_matrix)[0,1])

0.37997836

from sklearn.metrics.pairwise import pairwise_distances
print(pairwise_distances(tfidf_matrix[0:1], tfidf_matrix, metric='cosine')[0,1])

0.62002164

为什么这些是不同的？

Answer 1

来自源代码documentation

Cosine distance is defined as 1.0 minus the cosine similarity.

所以你的结果是有意义的。

Answer 2

成对距离提供两个阵列之间的距离更多的成对距离意味着更小的similarity.while余弦相似度是1-成对距离，所以更多的余弦相似度意味着两个阵列之间的相似度更高。