Python：tf-idf-cosine：查找文档相似性

时间：2012-08-25 02:41:26

标签： python machine-learning nltk information-retrieval tf-idf

我正在关注Part 1＆amp; Part 2。不幸的是，作者没有时间进行涉及使用余弦相似性的最后一节实际找到两个文档之间的距离。我在stackoverflow的以下链接的帮助下跟踪了文章中的示例，其中包括上面链接中提到的代码（只是为了让生活更轻松）

from sklearn.feature_extraction.text import CountVectorizer
from sklearn.feature_extraction.text import TfidfTransformer
from nltk.corpus import stopwords
import numpy as np
import numpy.linalg as LA

train_set = ["The sky is blue.", "The sun is bright."]  # Documents
test_set = ["The sun in the sky is bright."]  # Query
stopWords = stopwords.words('english')

vectorizer = CountVectorizer(stop_words = stopWords)
#print vectorizer
transformer = TfidfTransformer()
#print transformer

trainVectorizerArray = vectorizer.fit_transform(train_set).toarray()
testVectorizerArray = vectorizer.transform(test_set).toarray()
print 'Fit Vectorizer to train set', trainVectorizerArray
print 'Transform Vectorizer to test set', testVectorizerArray

transformer.fit(trainVectorizerArray)
print
print transformer.transform(trainVectorizerArray).toarray()

transformer.fit(testVectorizerArray)
print 
tfidf = transformer.transform(testVectorizerArray)
print tfidf.todense()

作为上述代码的结果，我有以下矩阵

Fit Vectorizer to train set [[1 0 1 0]
 [0 1 0 1]]
Transform Vectorizer to test set [[0 1 1 1]]

[[ 0.70710678  0.          0.70710678  0.        ]
 [ 0.          0.70710678  0.          0.70710678]]

[[ 0.          0.57735027  0.57735027  0.57735027]]

我不知道如何使用此输出来计算余弦相似度，我知道如何对两个相似长度的矢量实现余弦相似度，但在这里我不知道如何识别这两个矢量。

6 个答案:

答案 0 :(得分：138)

首先，如果你想提取计数特征并应用TF-IDF规范化和逐行欧几里德规范化，你可以使用TfidfVectorizer进行一次操作：

>>> from sklearn.feature_extraction.text import TfidfVectorizer
>>> from sklearn.datasets import fetch_20newsgroups
>>> twenty = fetch_20newsgroups()

>>> tfidf = TfidfVectorizer().fit_transform(twenty.data)
>>> tfidf
<11314x130088 sparse matrix of type '<type 'numpy.float64'>'
    with 1787553 stored elements in Compressed Sparse Row format>

现在找到一个文档的余弦距离（例如数据集中的第一个）和所有其他文档，你只需要计算第一个向量的点积和所有其他文件，因为tfidf向量已经是行 - 标准化。 scipy稀疏矩阵API有点奇怪（不像密集的N维numpy数组那样灵活）。要获得第一个向量，您需要逐行切割矩阵以获得具有单行的子矩阵：

>>> tfidf[0:1]
<1x130088 sparse matrix of type '<type 'numpy.float64'>'
    with 89 stored elements in Compressed Sparse Row format>

scikit-learn已经提供了成对指标（机器学习用语中的a.k.a.内核），它适用于向量集合的密集和稀疏表示。在这种情况下，我们需要一个点积，也称为线性内核：

>>> from sklearn.metrics.pairwise import linear_kernel
>>> cosine_similarities = linear_kernel(tfidf[0:1], tfidf).flatten()
>>> cosine_similarities
array([ 1.        ,  0.04405952,  0.11016969, ...,  0.04433602,
    0.04457106,  0.03293218])

因此，为了找到前5个相关文档，我们可以使用argsort和一些负数组切片（大多数相关文档具有最高余弦相似度值，因此在排序索引数组的末尾）：

>>> related_docs_indices = cosine_similarities.argsort()[:-5:-1]
>>> related_docs_indices
array([    0,   958, 10576,  3277])
>>> cosine_similarities[related_docs_indices]
array([ 1.        ,  0.54967926,  0.32902194,  0.2825788 ])

第一个结果是健全性检查：我们发现查询文档是最相似的文档，余弦相似度得分为1，其中包含以下文本：

>>> print twenty.data[0]
From: lerxst@wam.umd.edu (where's my thing)
Subject: WHAT car is this!?
Nntp-Posting-Host: rac3.wam.umd.edu
Organization: University of Maryland, College Park
Lines: 15

 I was wondering if anyone out there could enlighten me on this car I saw
the other day. It was a 2-door sports car, looked to be from the late 60s/
early 70s. It was called a Bricklin. The doors were really small. In addition,
the front bumper was separate from the rest of the body. This is
all I know. If anyone can tellme a model name, engine specs, years
of production, where this car is made, history, or whatever info you
have on this funky looking car, please e-mail.

Thanks,
- IL
   ---- brought to you by your neighborhood Lerxst ----

第二个最相似的文档是引用原始邮件的回复，因此有许多常用词：

>>> print twenty.data[958]
From: rseymour@reed.edu (Robert Seymour)
Subject: Re: WHAT car is this!?
Article-I.D.: reed.1993Apr21.032905.29286
Reply-To: rseymour@reed.edu
Organization: Reed College, Portland, OR
Lines: 26

In article <1993Apr20.174246.14375@wam.umd.edu> lerxst@wam.umd.edu (where's my
thing) writes:
>
>  I was wondering if anyone out there could enlighten me on this car I saw
> the other day. It was a 2-door sports car, looked to be from the late 60s/
> early 70s. It was called a Bricklin. The doors were really small. In
addition,
> the front bumper was separate from the rest of the body. This is
> all I know. If anyone can tellme a model name, engine specs, years
> of production, where this car is made, history, or whatever info you
> have on this funky looking car, please e-mail.

Bricklins were manufactured in the 70s with engines from Ford. They are rather
odd looking with the encased front bumper. There aren't a lot of them around,
but Hemmings (Motor News) ususally has ten or so listed. Basically, they are a
performance Ford with new styling slapped on top.

>    ---- brought to you by your neighborhood Lerxst ----

Rush fan?

--
Robert Seymour              rseymour@reed.edu
Physics and Philosophy, Reed College    (NeXTmail accepted)
Artificial Life Project         Reed College
Reed Solar Energy Project (SolTrain)    Portland, OR

答案 1 :(得分：18)

我知道这是一个老帖子。但是我尝试了http://scikit-learn.sourceforge.net/stable/包。这是我的代码，以找到余弦相似性。问题是如何计算与此包的余弦相似度，这是我的代码

from sklearn.feature_extraction.text import CountVectorizer
from sklearn.metrics.pairwise import cosine_similarity
from sklearn.feature_extraction.text import TfidfVectorizer

f = open("/root/Myfolder/scoringDocuments/doc1")
doc1 = str.decode(f.read(), "UTF-8", "ignore")
f = open("/root/Myfolder/scoringDocuments/doc2")
doc2 = str.decode(f.read(), "UTF-8", "ignore")
f = open("/root/Myfolder/scoringDocuments/doc3")
doc3 = str.decode(f.read(), "UTF-8", "ignore")

train_set = ["president of India",doc1, doc2, doc3]

tfidf_vectorizer = TfidfVectorizer()
tfidf_matrix_train = tfidf_vectorizer.fit_transform(train_set)  #finds the tfidf score with normalization
print "cosine scores ==> ",cosine_similarity(tfidf_matrix_train[0:1], tfidf_matrix_train)  #here the first element of tfidf_matrix_train is matched with other three elements

这里假设查询是train_set的第一个元素，doc1，doc2和doc3是我想借助余弦相似性排序的文档。然后我可以使用这段代码。

此问题中提供的教程非常有用。以下是它的所有部分 part-I，part-II，part-III

输出如下：

[[ 1.          0.07102631  0.02731343  0.06348799]]

这里1表示查询与自身匹配，其他三个是将查询与相应文档匹配的分数。

答案 2 :(得分：17)

在@ excray评论的帮助下，我设法找出答案，我们需要做的是实际编写一个简单的for循环来迭代表示列车数据和测试数据的两个数组。

首先实现一个简单的lambda函数来保存余弦计算的公式：

cosine_function = lambda a, b : round(np.inner(a, b)/(LA.norm(a)*LA.norm(b)), 3)

然后只需编写一个简单的for循环来遍历to vector，逻辑就是每个“对于trainVectorizerArray中的每个向量，你必须在testVectorizerArray中找到与向量的余弦相似性。”

from sklearn.feature_extraction.text import CountVectorizer
from sklearn.feature_extraction.text import TfidfTransformer
from nltk.corpus import stopwords
import numpy as np
import numpy.linalg as LA

train_set = ["The sky is blue.", "The sun is bright."] #Documents
test_set = ["The sun in the sky is bright."] #Query
stopWords = stopwords.words('english')

vectorizer = CountVectorizer(stop_words = stopWords)
#print vectorizer
transformer = TfidfTransformer()
#print transformer

trainVectorizerArray = vectorizer.fit_transform(train_set).toarray()
testVectorizerArray = vectorizer.transform(test_set).toarray()
print 'Fit Vectorizer to train set', trainVectorizerArray
print 'Transform Vectorizer to test set', testVectorizerArray
cx = lambda a, b : round(np.inner(a, b)/(LA.norm(a)*LA.norm(b)), 3)

for vector in trainVectorizerArray:
    print vector
    for testV in testVectorizerArray:
        print testV
        cosine = cx(vector, testV)
        print cosine

transformer.fit(trainVectorizerArray)
print
print transformer.transform(trainVectorizerArray).toarray()

transformer.fit(testVectorizerArray)
print 
tfidf = transformer.transform(testVectorizerArray)
print tfidf.todense()

这是输出：

Fit Vectorizer to train set [[1 0 1 0]
 [0 1 0 1]]
Transform Vectorizer to test set [[0 1 1 1]]
[1 0 1 0]
[0 1 1 1]
0.408
[0 1 0 1]
[0 1 1 1]
0.816

[[ 0.70710678  0.          0.70710678  0.        ]
 [ 0.          0.70710678  0.          0.70710678]]

[[ 0.          0.57735027  0.57735027  0.57735027]]

答案 3 :(得分：16)

让我给你一个由我写的另一个教程。它回答了你的问题，但也解释了我们为什么要做一些事情。我也试着简明扼要。

所以你有一个list_of_documents只是一个字符串数组，另一个document只是一个字符串。您需要从与list_of_documents最相似的document中找到此类文档。

让我们将它们组合在一起：documents = list_of_documents + [document]

让我们从依赖开始。很明显我们为什么要使用它们。

from nltk.corpus import stopwords
import string
from nltk.tokenize import wordpunct_tokenize as tokenize
from nltk.stem.porter import PorterStemmer
from sklearn.feature_extraction.text import TfidfVectorizer
from scipy.spatial.distance import cosine

可以使用的方法之一是bag-of-words方法，我们将文档中的每个单词独立于其他单词处理，然后将所有单词放在大包中。从一个角度来看，它丢失了很多信息（比如单词是如何连接的），但从另一个角度来看，它使模型变得简单。

在英语和任何其他人类语言中，有很多“无用的”词语，如“a”，“the”，“in”，这些词很常见，不具备很多意义。它们被称为stop words，删除它们是个好主意。人们可以注意到的另一件事是“分析”，“分析器”，“分析”等词语非常相似。它们有一个共同的根，所有都可以转换为一个单词。这个过程称为stemming，并且存在不同的词干分析器，它们在速度，攻击性等方面不同。因此，我们将每个文档转换为单词的词干列表，没有停用词。我们也会丢弃所有标点符号。

porter = PorterStemmer()
stop_words = set(stopwords.words('english'))

modified_arr = [[porter.stem(i.lower()) for i in tokenize(d.translate(None, string.punctuation)) if i.lower() not in stop_words] for d in documents]

那么这一句话怎么会帮助我们呢？想象一下，我们有3个包：[a, b, c]，[a, c, a]和[b, c, d]。我们可以将它们转换为vectors in the basis [a, b, c, d]。因此，我们最终得到了向量：[1, 1, 1, 0]，[2, 0, 1, 0]和[0, 1, 1, 1]。类似的事情是我们的文件（只有矢量会更长）。现在我们看到我们删除了很多单词并阻止了其他单词也减少了向量的维数。这里有一个有趣的观察。较长的文档将比较短的文档具有更多的正面元素，这就是为什么将矢量规范化很好。这被称为术语频率TF，人们还使用关于该词在其他文档中使用频率的附加信息 - 逆文档频率IDF。我们一起有一个指标TF-IDF which have a couple of flavors。这可以通过sklearn中的一行来实现:-) 

modified_doc = [' '.join(i) for i in modified_arr] # this is only to convert our list of lists to list of strings that vectorizer uses.
tf_idf = TfidfVectorizer().fit_transform(modified_doc)

实际上，矢量化程序allows to do a lot of things就像删除停用词和小写一样。我之所以在一个单独的步骤中完成它们只是因为sklearn没有非英语的停用词，但是nltk有。

所以我们计算了所有的向量。最后一步是找出哪一个与最后一个最相似。有多种方法可以实现这一点，其中一种方法是欧几里德距离，因为discussed here的原因并不是很大。另一种方法是cosine similarity。我们迭代所有文档并计算文档与最后一个文档之间的余弦相似度：

l = len(documents) - 1
for i in xrange(l):
    minimum = (1, None)
    minimum = min((cosine(tf_idf[i].todense(), tf_idf[l + 1].todense()), i), minimum)
print minimum

现在最小值将包含有关最佳文档及其分数的信息。

答案 4 :(得分：12)

这应该对你有所帮助。 

from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.metrics.pairwise import cosine_similarity  

tfidf_vectorizer = TfidfVectorizer()
tfidf_matrix = tfidf_vectorizer.fit_transform(train_set)
print tfidf_matrix
cosine = cosine_similarity(tfidf_matrix[length-1], tfidf_matrix)
print cosine

并输出：

[[ 0.34949812  0.81649658  1.        ]]

答案 5 :(得分：1)

这里是将测试数据与训练数据进行比较的功能，其中装有训练数据的Tf-Idf变压器。优点是您可以快速地进行透视或分组以找到n个最接近的元素，并且计算是按矩阵方式进行的。

def create_tokenizer_score(new_series, train_series, tokenizer):
    """
    return the tf idf score of each possible pairs of documents
    Args:
        new_series (pd.Series): new data (To compare against train data)
        train_series (pd.Series): train data (To fit the tf-idf transformer)
    Returns:
        pd.DataFrame
    """

    train_tfidf = tokenizer.fit_transform(train_series)
    new_tfidf = tokenizer.transform(new_series)
    X = pd.DataFrame(cosine_similarity(new_tfidf, train_tfidf), columns=train_series.index)
    X['ix_new'] = new_series.index
    score = pd.melt(
        X,
        id_vars='ix_new',
        var_name='ix_train',
        value_name='score'
    )
    return score

train_set = pd.Series(["The sky is blue.", "The sun is bright."])
test_set = pd.Series(["The sun in the sky is bright."])
tokenizer = TfidfVectorizer() # initiate here your own tokenizer (TfidfVectorizer, CountVectorizer, with stopwords...)
score = create_tokenizer_score(train_series=train_set, new_series=test_set, tokenizer=tokenizer)
score

   ix_new   ix_train    score
0   0       0       0.617034
1   0       1       0.862012
相关问题
最新问题