FreqDist用于最常见的单词或短语

Question

I'm trying to analyze some data from app reviews.

I want to use nltk's FreqDist to see the most frequently occurring phrases in a file. It can be a single token or key phrases. I don't want to tokenize the data because that would give me most frequent tokens only. But right now, the FreqDist function is processing each review as one string, and is not extracting the words in each review.

df = pd.read_csv('Positive.csv')

def pre_process(text):
    translator = str.maketrans("", "", string.punctuation)
    text = text.lower().strip().replace("\n", " ").replace("’", "").translate(translator)
    return text

df['Description'] = df['Description'].map(pre_process)
df = df[df['Description'] != '']

word_dist = nltk.FreqDist(df['Description'])

('Description' is the body/message of the reviews.)

For example, I want to get something like Most Frequent terms: "I like", "useful", "very good app" But instead I'm getting Most Frequent terms: "I really enjoy this app because bablabla" (entire review)

And that's why when I'm plotting the FreqDist I get this:

Answer 1

TL;DR

Use ngrams or everygrams :

>>> from itertools import chain
>>> import pandas as pd
>>> from nltk import word_tokenize
>>> from nltk import FreqDist

>>> df = pd.read_csv('x')
>>> df['Description']
0            Here is a sentence.
1    This is a foo bar sentence.
Name: Description, dtype: object

>>> df['Description'].map(word_tokenize)
0              [Here, is, a, sentence, .]
1    [This, is, a, foo, bar, sentence, .]
Name: Description, dtype: object

>>> sents = df['Description'].map(word_tokenize).tolist()

>>> FreqDist(list(chain(*[everygrams(sent, 1, 3) for sent in sents])))
FreqDist({('sentence',): 2, ('is', 'a'): 2, ('sentence', '.'): 2, ('is',): 2, ('.',): 2, ('a',): 2, ('Here', 'is', 'a'): 1, ('a', 'foo'): 1, ('a', 'sentence'): 1, ('bar', 'sentence', '.'): 1, ...})