我尝试使用公式手动计算 tfidf 值，但得到的结果与使用 scikit-learn 实现时得到的结果不同。
from sklearn.feature_extraction.text import TfidfVectorizer
tv = TfidfVectorizer()
a = "cat hat bat splat cat bat hat mat cat"
b = "cat mat cat sat"
tv.fit_transform([a, b]).toarray()
# array([[0.53333448, 0.56920781, 0.53333448, 0.18973594, 0. ,
# 0.26666724],
# [0. , 0.75726441, 0. , 0.37863221, 0.53215436,
# 0. ]])
tv.get_feature_names()
# ['bat', 'cat', 'hat', 'mat', 'sat', 'splat']
我尝试手动计算文档的
tfidf 但结果与 TfidfVectorizer.fit_transform 不同。(np.log(2+1/1+1) + 1) * (2/9) = 0.5302876358044202
(np.log(2+1/2+1) + 1) * (3/9) = 0.750920989498456
(np.log(2+1/1+1) + 1) * (2/9) = 0.5302876358044202
(np.log(2+1/2+1) + 1) * (1/9) = 0.25030699649948535
(np.log(2+1/1+1) + 1) * (0/9) = 0.0
(np.log(2+1/1+1) + 1) * (1/9) = 0.2651438179022101
我应该得到的是
[0.53333448, 0.56920781, 0.53333448, 0.18973594, 0, 0.26666724]
最佳答案
TFIDF 有许多变体。 sklearn使用的公式是:
(count_of_term_t_in_d) * ((log ((NUMBER_OF_DOCUMENTS + 1) / (Number_of_documents_where_t_appears +1 )) + 1)
2 * (np.log((1 + 2)/(1+1)) + 1) = 2.8109302162163288
3 * (np.log((1 + 2)/(2+1)) + 1) = 3.0
2 * (np.log((1 + 2)/(1+1)) + 1) = 2.8109302162163288
1 * (np.log((1 + 2)/(2+1)) + 1) = 1.0
0 * (np.log((1 + 2)/(2+1)) + 1) = 0.0
1 * (np.log((1 + 2)/(1+1)) + 1) = 1.4054651081081644
经过计算，最终的TFIDF向量通过欧几里德范数归一化:
tfidf_vector = [2.8109302162163288, 3.0, 2.8109302162163288, 1.0, 0.0, 1.4054651081081644]
tfidf_vector = tfidf_vector / np.linalg.norm(tfidf_vector)
print(tfidf_vector)
[0.53333448, 0.56920781, 0.53333448, 0.18973594, 0, 0.26666724]
关于python - scikit 学习 tfidf 的实现与手动实现不同，我们在Stack Overflow上找到一个类似的问题：https://stackoverflow.com/questions/54746926/