41

I am new to scikit-learn, and I was using TfidfVectorizer to find the tfidf values of terms in a set of documents. I used the following code to obtain the same.

vectorizer = TfidfVectorizer(stop_words=u'english',ngram_range=(1,5),lowercase=True)
X = vectorizer.fit_transform(lectures)

Now If I print X, I am able to see all the entries in matrix, but how can I find top n entries based on tfidf score. In addition to that is there any method that will help me to find top n entries based on tfidf score per ngram i.e. top entries among unigram,bigram,trigram and so on?

4

1 回答 1

64

从 0.15 版本开始，a 学习到的特征的全局项权重TfidfVectorizer可以通过属性访问idf_，这将返回一个长度等于特征维度的数组。按此权重对特征进行排序以获得权重最高的特征：

from sklearn.feature_extraction.text import TfidfVectorizer
import numpy as np

lectures = ["this is some food", "this is some drink"]
vectorizer = TfidfVectorizer()
X = vectorizer.fit_transform(lectures)
indices = np.argsort(vectorizer.idf_)[::-1]
features = vectorizer.get_feature_names()
top_n = 2
top_features = [features[i] for i in indices[:top_n]]
print top_features

输出：

[u'food', u'drink']

通过 ngram 获取顶级特征的第二个问题可以使用相同的想法来完成，但需要一些额外的步骤将特征分成不同的组：

from sklearn.feature_extraction.text import TfidfVectorizer
from collections import defaultdict

lectures = ["this is some food", "this is some drink"]
vectorizer = TfidfVectorizer(ngram_range=(1,2))
X = vectorizer.fit_transform(lectures)
features_by_gram = defaultdict(list)
for f, w in zip(vectorizer.get_feature_names(), vectorizer.idf_):
    features_by_gram[len(f.split(' '))].append((f, w))
top_n = 2
for gram, features in features_by_gram.iteritems():
    top_features = sorted(features, key=lambda x: x[1], reverse=True)[:top_n]
    top_features = [f[0] for f in top_features]
    print '{}-gram top:'.format(gram), top_features

输出：

1-gram top: [u'drink', u'food']
2-gram top: [u'some drink', u'some food']
于 2014-08-09T14:24:09.407 回答