Scikit Learn TfidfVectorizer : How to get top n terms with highest tf-idf score

Question

I am working on keyword extraction problem. Consider the very general case

from sklearn.feature_extraction.text import TfidfVectorizer

tfidf = TfidfVectorizer(tokenizer=tokenize, stop_words='english')

t = """Two Travellers, walking in the noonday sun, sought the shade of a widespreading tree to rest. As they lay looking up among the pleasant leaves, they saw that it was a Plane Tree.

"How useless is the Plane!" said one of them. "It bears no fruit whatever, and only serves to litter the ground with leaves."

"Ungrateful creatures!" said a voice from the Plane Tree. "You lie here in my cooling shade, and yet you say I am useless! Thus ungratefully, O Jupiter, do men receive their blessings!"

Our best blessings are often the least appreciated."""

tfs = tfidf.fit_transform(t.split(" "))
str = 'tree cat travellers fruit jupiter'
response = tfidf.transform([str])
feature_names = tfidf.get_feature_names()

for col in response.nonzero()[1]:
    print(feature_names[col], ' - ', response[0, col])

and this gives me

  (0, 28)   0.443509712811
  (0, 27)   0.517461475101
  (0, 8)    0.517461475101
  (0, 6)    0.517461475101
tree  -  0.443509712811
travellers  -  0.517461475101
jupiter  -  0.517461475101
fruit  -  0.517461475101

which is good. For any new document that comes in, is there a way to get the top n terms with the highest tfidf score?

Solution 1

You have to do a little bit of a song and dance to get the matrices as numpy arrays instead, but this should do what you're looking for:

feature_array = np.array(tfidf.get_feature_names())
tfidf_sorting = np.argsort(response.toarray()).flatten()[::-1]

n = 3
top_n = feature_array[tfidf_sorting][:n]

This gives me:

array([u'fruit', u'travellers', u'jupiter'], 
  dtype='<U13')

The argsort call is really the useful one, here are the docs for it. We have to do [::-1] because argsort only supports sorting small to large. We call flatten to reduce the dimensions to 1d so that the sorted indices can be used to index the 1d feature array. Note that including the call to flatten will only work if you're testing one document at at time.

Also, on another note, did you mean something like tfs = tfidf.fit_transform(t.split("\n\n"))? Otherwise, each term in the multiline string is being treated as a "document". Using \n\n instead means that we are actually looking at 4 documents (one for each line), which makes more sense when you think about tfidf.

Solution 2

Solution using sparse matrix itself (without .toarray())!

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

tfidf = TfidfVectorizer(stop_words='english')
corpus = [
    'I would like to check this document',
    'How about one more document',
    'Aim is to capture the key words from the corpus',
    'frequency of words in a document is called term frequency'
]

X = tfidf.fit_transform(corpus)
feature_names = np.array(tfidf.get_feature_names())


new_doc = ['can key words in this new document be identified?',
           'idf is the inverse document frequency caculcated for each of the words']
responses = tfidf.transform(new_doc)


def get_top_tf_idf_words(response, top_n=2):
    sorted_nzs = np.argsort(response.data)[:-(top_n+1):-1]
    return feature_names[response.indices[sorted_nzs]]
  
print([get_top_tf_idf_words(response,2) for response in responses])

#[array(['key', 'words'], dtype='<U9'),
 array(['frequency', 'words'], dtype='<U9')]

Solution 3

Here is a quick code for that: (documents is a list)

def get_tfidf_top_features(documents,n_top=10):
  fidf_vectorizer = TfidfVectorizer(max_df=0.95, min_df=2, max_features=no_features, stop_words='english')
  tfidf = tfidf_vectorizer.fit_transform(documents)
  importance = np.argsort(np.asarray(tfidf.sum(axis=0)).ravel())[::-1]
  tfidf_feature_names = np.array(tfidf_vectorizer.get_feature_names())
  return tfidf_feature_names[importance[:n_top]]

Comments

[iamdeit]

How would I achieve that by using DictVectorizer + TfidfTransformer?

[Pedram]

What if we want to list top n terms for each class not for each document? I asked a question here but no response yet!

[hume]

What do you mean, "for each class". Say you have documents labeled "A" and documents labeled "B". You could either: (1) calculate "TF-ICF", which gives you the term frequency-inverse class frequency. Do that just by concatenating all of the documents of a class into a single string and doing the normal tfidf process. (2) Alternatively, you could calculate average TFIDF, by adding the class as a column to the TFIDF, creating a pandas dataframe, and then doing tfidf_dataframe.groupby('class_name').mean().

[function]

Strangely, The last line gives memory errors , while replacing it to top_n = feature_array[tfidf_sorting[:n]] it doesn't .

[Outcast]

@Pedram, I asked the same question (stackoverflow.com/questions/56703244/…) for per class you mention it at your comment above. Do you have an answer to it?

[Outcast]

By the way, @hume this line tfidf_sorting = np.argsort(response.toarray()).flatten()[::-1] gives me a memory error which must be because my tf-idf matrix is too big. So I guess that I could do this in batches?

[Atlas]

I haven't looked into this at all, but casting tfidf.get_feature_names() as an numpy.array uses massively more memory than the default Python list. My 300mb TFIDF model turns into 4+ Gb in RAM when I call numpy.array on get_feature_names(), whereas simply using feature_array = tfidf.get_feature_names() works fine and uses very little RAM.

[Akash Singh]

It returns the repetitive words also, When I am trying to use these top n words as my vocabulary in tfidfvectorizer again, it throws and value error with as there are duplicate words in vocab. How will I get top n unique words?

[Venkatachalam]

Interesting. I am using get_feature_names() to get the feature_names, hence there should not be any duplicates returned by get_top_tf_idf_words. Can you post a new question, with a reproducible example and tag me?

[raj]

@Atlas feature_array = tfidf.get_feature_names() worked for me

[aoez]

There is a typo in the second line. The first character "t" is missing.

[SriK]

no_features is missing variable.