-
Notifications
You must be signed in to change notification settings - Fork 0
/
nlp.py
51 lines (36 loc) · 1.48 KB
/
nlp.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
import nltk
import yaml
from bs4 import BeautifulSoup
from langdetect import detect
from nltk.corpus import stopwords
from nltk.tokenize import RegexpTokenizer
from textblob import TextBlob
config = yaml.safe_load(open('config.yml'))
tokenizer = RegexpTokenizer(r'\w+')
def _text_get_lang(text):
return detect(text)
def _text_get_nouns(text):
# TODO: Try replacing with nltk as textblob's performance (especially on German) isn't that good..
return TextBlob(text).noun_phrases
def _clear_html(content):
return BeautifulSoup(content, 'lxml').text
def _remove_stopwords(tokens, language='english'):
return list(filter(lambda word: (word not in stopwords.words(language)
and str(word).isalpha()) and len(word) > config['processing']['min_word_length'], tokens))
def _str_get_tokens(text):
return [token.lower() for token in tokenizer.tokenize(text)]
def _get_word_frequencies(tokens):
return nltk.FreqDist(tokens)
def nlp_process(content, clear_html=True, extract_keywords=True, summerize=True):
return [], []
# content_cleared = _clear_html(content) if clear_html else content
#
# nouns = _text_get_nouns(content_cleared)
#
# tokens = _str_get_tokens(content_cleared)
# filtered_tokens = _remove_stopwords(tokens)
# word_frequencies = _get_word_frequencies(filtered_tokens)
#
# # TODO: Only use nouns as keywords
#
# return word_frequencies.most_common(15) if extract_keywords else None, ''