-
Notifications
You must be signed in to change notification settings - Fork 0
/
lda_baseline.py
49 lines (42 loc) · 1.73 KB
/
lda_baseline.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
import pandas as pd
# baseline implementation of LDA for comparison purposes in future
from urllib.parse import urlparse
import ast
import numpy as np
from nltk.corpus import stopwords
import gensim
from gensim.utils import simple_preprocess
from gensim import corpora
stop_words = stopwords.words('english')
stop_words.extend(['from', 'subject', 're', 'edu', 'use'])
# method to remove stopwords
def remove_stopwords(texts):
return [[word for word in simple_preprocess(str(doc)) if word not in stop_words] for doc in texts]
# method to bigrams
def make_bigrams(texts,bigram_mod):
return [bigram_mod[doc] for doc in texts]
# method for making tri-grams
def make_trigrams(texts,bigram_mod,trigram_mod):
return [trigram_mod[bigram_mod[doc]] for doc in texts]
# method for sentence to words generation
def sent_to_words(sentences):
for sentence in sentences:
yield(gensim.utils.simple_preprocess(str(sentence), deacc=True))
# function to compute topics
def lda_complete(text,bigram_min_count,threshold,num_topics):
"""
params:
1. text - list of words
2. bigram_min_count - minimum frequency to form a bigram
3. threshold - max number of bigrams
4. num_topics - number of topics
"""
bigram = gensim.models.Phrases(text, min_count=bigram_min_count, threshold=threshold)
trigram = gensim.models.Phrases(bigram[text], threshold=threshold)
bigram_mod = gensim.models.phrases.Phraser(bigram)
trigram_mod = gensim.models.phrases.Phraser(trigram)
texts = make_bigrams(remove_stopwords(data_words),bigram_mod)
id2word = corpora.Dictionary(texts)
corpus = [id2word.doc2bow(text) for text in texts]
lda_model = gensim.models.ldamodel.LdaModel(corpus=corpus,id2word=id2word,num_topics=num_topics)
return print(lda_model.print_topics())