Blog

Text analysis in Pandas with some TF-IDF (again)

Posted By Jakub Nowacki, 18 September 2017

Pandas is a great tool for the analysis of tabular data via its DataFrame interface. Slightly less known are its capabilities for working with text data. In this post I’ll present them on some simple examples. As a comparison I’ll use my previous post about TF-IDF in Spark.

# Some used imports
%matplotlib inline
import pandas as pd
import numpy as np
import os
import glob
import matplotlib as mpl

# Just making the plots look better
mpl.style.use('ggplot')
mpl.rcParams['figure.figsize'] = (8,6)
mpl.rcParams['font.size'] = 12

First we load data, i.e. books previously downloaded from Project Gutenberg site. Here we use a handy glob module to walk over text files in a directory and read in files line by line into DataFrame.

books = glob.glob('data/*.txt')
d = list()
for book_file in books:
    with open(book_file, encoding='utf-8') as f:
        book = os.path.basename(book_file.split('.')[0])
        d.append(pd.DataFrame({'book': book, 'lines': f.readlines()}))
doc = pd.concat(d)
doc.head()
book lines
0 dracula The Project Gutenberg EBook of Dracula, by Br...
1 dracula \n
2 dracula This eBook is for the use of anyone anywhere a...
3 dracula almost no restrictions whatsoever. You may co...
4 dracula re-use it under the terms of the Project Guten...

Since we’re now in Pandas, we can easily use its plotting capability to look at the number of lines in the books. Note that value_counts of books create Series indexed by the book names, thus, we don’t need to set index for plotting. Note that we can now choose plot styling; see demo page for available styles.

doc['book'].value_counts().plot.bar();

png

Now, we need to tokenize the sentences into words aka terms. While we can do it in a loop, we can take advantage of the split function in the text toolkit for Pandas’ Series; see this manual for all the functions. To get it we just invoke the strip function, which is a part of str, i.e. StringsMethods object. The argument is regular expression pattern, on which the string, in our case the sentence, will be split. In our case it is the regular expression set of everything that is not a letter (capital or not) or digit and underscore. This is because \W is the reverse of \w which is a set [A-Za-z0-9_], i.e. contains an underscore. In this case we want also to split on underscore so we had to add it to the splitting set. Note that this decision and other you’ll make while deciding on the splitting set, will influence the tokenization, and thus the result. Hence, select the splitting pattern carefully. If you’re not sure about your decision, use practicing pages, like regex101.com to check your patterns.

doc['words'] = doc.lines.str.strip().str.split('[\W_]+')
doc.head()
book lines words
0 dracula The Project Gutenberg EBook of Dracula, by Br... [, The, Project, Gutenberg, EBook, of, Dracula...
1 dracula \n []
2 dracula This eBook is for the use of anyone anywhere a... [This, eBook, is, for, the, use, of, anyone, a...
3 dracula almost no restrictions whatsoever. You may co... [almost, no, restrictions, whatsoever, You, ma...
4 dracula re-use it under the terms of the Project Guten... [re, use, it, under, the, terms, of, the, Proj...

As a result we get a new column with array of words. Now we need to flatten the resulted DataFrame somehow. The best approach is to use iterators and create new DataFrame with words; see this StackOverflow post for the performance details of this solution. Note that this may take a few seconds depending on the machine; on mine it takes about 25s.

rows = list()
for row in doc[['book', 'words']].iterrows():
    r = row[1]
    for word in r.words:
        rows.append((r.book, word))

words = pd.DataFrame(rows, columns=['book', 'word'])
words.head()
book word
0 dracula
1 dracula The
2 dracula Project
3 dracula Gutenberg
4 dracula EBook

Now we have DataFrame with words, but occasionally we get an empty string as a byproduct of the splitting. We should remove it as it would be counted as a term. To do that we can use function len as follows:

words = words[words.word.str.len() > 0]
words.head()
book word
1 dracula The
2 dracula Project
3 dracula Gutenberg
4 dracula EBook
5 dracula of

The new words DataFrame is now free of the empty strings.

If we want to calculate TF-IDF statistic we need to normalize the words by bringing them all to the same case. Again, we can use a Series string function for that.

words['word'] = words.word.str.lower()
words.head()
book word
1 dracula the
2 dracula project
3 dracula gutenberg
4 dracula ebook
5 dracula of

First, lets calculate counts of the terms per book. We can do it as follows:

counts = words.groupby('book')\
    .word.value_counts()\
    .to_frame()\
    .rename(columns={'word':'n_w'})
counts.head()
n_w
book word
dracula the 8093
and 5976
i 4847
to 4745
of 3748

Note that as a result we are getting a hierarchical index aka MultiIndex; see Pandas manual for more details.

With this index we can now plot the results in much nicer way. E.e. we can get 5 words largest counts per book and plot it as shown below. Note that we need to reset, i.e. remove one level of index, as it doubles when we are using nlargest function. I’ve made the process into a function, as I will reuse it further.

def pretty_plot_top_n(series, top_n=5, index_level=0):
    r = series\
    .groupby(level=index_level)\
    .nlargest(top_n)\
    .reset_index(level=index_level, drop=True)
    r.plot.bar()
    return r.to_frame()


pretty_plot_top_n(counts['n_w'])
n_w
book word
dracula the 8093
and 5976
i 4847
to 4745
of 3748
frankenstein the 4371
and 3046
i 2850
of 2760
to 2174
grimms_fairy_tales the 7224
and 5551
to 2749
he 2096
a 1978
moby_dick the 14718
of 6743
and 6518
a 4807
to 4707
tom_sawyer the 3973
and 3193
a 1955
to 1807
of 1585
war_and_peace the 34725
and 22307
to 16755
of 15008
a 10584

png

To finish the TF as shown in the previous post, we need the counts of the books to get the measure. As a reminder, below is the equation for TF:

Note that I’ve renamed the column as it inherits the name by default.

word_sum = counts.groupby(level=0)\
    .sum()\
    .rename(columns={'n_w': 'n_d'})
word_sum
n_d
book
dracula 166916
frankenstein 78475
grimms_fairy_tales 105336
moby_dick 222630
tom_sawyer 77612
war_and_peace 576627

Now we need to join both columns on book to get the sum of word per book into final DataFrame. Having both n_w and n_d, we can easily calculate TF as follows:

tf = counts.join(word_sum)

tf['tf'] = tf.n_w/tf.n_d

tf.head()
n_w n_d tf
book word
dracula the 8093 166916 0.048485
and 5976 166916 0.035802
i 4847 166916 0.029039
to 4745 166916 0.028427
of 3748 166916 0.022454

Again, lets look at the top 5 words with respect to TF.

pretty_plot_top_n(tf['tf'])
tf
book word
dracula the 0.048485
and 0.035802
i 0.029039
to 0.028427
of 0.022454
frankenstein the 0.055699
and 0.038815
i 0.036317
of 0.035170
to 0.027703
grimms_fairy_tales the 0.068581
and 0.052698
to 0.026097
he 0.019898
a 0.018778
moby_dick the 0.066110
of 0.030288
and 0.029277
a 0.021592
to 0.021143
tom_sawyer the 0.051191
and 0.041141
a 0.025189
to 0.023282
of 0.020422
war_and_peace the 0.060221
and 0.038685
to 0.029057
of 0.026027
a 0.018355

png

As before (and as expected), we’ve got the English stop-words.

Now we can do IDF; the remainder of the formula is given below:

First we need to get the document count. The simples solution is to use the Series’ method nunique, i.e. get size of set of unique elements in a series.

c_d = words.book.nunique()
c_d
6

Similarly, to get the number of unique books that every term appeared in, we can use the same method but on a grouped data. Again we need to rename the column. Note that sorting values is only for the presentation and it is not needed for the further computations.

idf = words.groupby('word')\
    .book\
    .nunique()\
    .to_frame()\
    .rename(columns={'book':'i_d'})\
    .sort_values('i_d')
idf.head()
i_d
word
laplandish 1
moluccas 1
molten 1
molliten 1
molière 1

Having all the components of the IDF formula, we can now calculate it as a new column.

idf['idf'] = np.log(c_d/idf.i_d.values)

idf.head()
i_d idf
word
laplandish 1 1.791759
moluccas 1 1.791759
molten 1 1.791759
molliten 1 1.791759
molière 1 1.791759

To get the final DataFrame we need to join both DataFrames as follows:

tf_idf = tf.join(idf)

tf_idf.head()
n_w n_d tf i_d idf
book word
dracula the 8093 166916 0.048485 6 0.0
and 5976 166916 0.035802 6 0.0
i 4847 166916 0.029039 6 0.0
to 4745 166916 0.028427 6 0.0
of 3748 166916 0.022454 6 0.0

Having now DataFrame with both TF and IDF values, we can calculate TF-IDF statistic.

tf_idf['tf_idf'] = tf_idf.tf * tf_idf.idf
tf_idf.head()
n_w n_d tf i_d idf tf_idf
book word
dracula the 8093 166916 0.048485 6 0.0 0.0
and 5976 166916 0.035802 6 0.0 0.0
i 4847 166916 0.029039 6 0.0 0.0
to 4745 166916 0.028427 6 0.0 0.0
of 3748 166916 0.022454 6 0.0 0.0

Again, lets see the top TF-IDF terms:

pretty_plot_top_n(tf_idf['tf_idf'])
tf_idf
book word
dracula helsing 0.003467
lucy 0.003231
mina 0.002619
van 0.002126
harker 0.001879
frankenstein clerval 0.001347
justine 0.001256
felix 0.001142
elizabeth 0.000813
frankenstein 0.000731
grimms_fairy_tales gretel 0.001667
hans 0.001304
tailor 0.001174
dwarf 0.001055
hansel 0.000799
moby_dick ahab 0.004161
whale 0.003873
whales 0.002181
stubb 0.002101
queequeg 0.002036
tom_sawyer huck 0.005956
tom 0.004305
becky 0.002655
joe 0.002406
sid 0.001847
war_and_peace pierre 0.006100
natásha 0.003769
rostóv 0.002411
prince 0.002318
moscow 0.002243

png

As before we’ve got a set of important words for the given document.

Note that this more of a demo of Pandas text processing. If you’re considering using TF-IDF in a more production example, see some existing solutions like scikit-learn’s TfidfVectorizer.

Note that I’ve just scratched a surface with the Pandas’ text processing capabilietes. There are many other useful functions like the match function shown below:

r = words[words.word.str.match('^s')]\
    .groupby('word')\
    .count()\
    .rename(columns={'book': 'n'})\
    .nlargest(10, 'n')
r.plot.bar()
r
n
word
s 8341
she 6317
so 5380
said 5337
some 2317
see 1760
such 1456
still 1368
should 1264
seemed 1233

png

Again, I encourage you to check Pandas manual for more details.

Notebook with the above computations is available for download here.

Contact

How can we help?

Drop us a line and we'll respond as soon as possible.

Treble