-
Notifications
You must be signed in to change notification settings - Fork 0
/
lda.py
132 lines (106 loc) · 3.93 KB
/
lda.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
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
from nltk.corpus import stopwords
from nltk.stem.wordnet import WordNetLemmatizer
import string, os, re, json, threading, gensim
import xml.etree.ElementTree as ET
from gensim import corpora
from scrape import getPMCXML
import config.config as CONFIG
stop = set(stopwords.words('english'))
exclude = set(string.punctuation)
lemma = WordNetLemmatizer()
docs = []
dictionary = {}
def clean(doc):
stop_free = " ".join([i for i in doc.lower().split() if i not in stop])
punc_free = ''.join(ch for ch in stop_free if ch not in exclude)
normalized = " ".join(lemma.lemmatize(word) for word in punc_free.split())
return normalized
class pubThread(threading.Thread):
def __init__(self, threadID, pmcids):
threading.Thread.__init__(self)
self.threadID = threadID
self.pmcids = pmcids
def run(self):
print('Starting thread', self.threadID)
global docs
for pmcid in self.pmcids:
raw_text = getPMCXML(pmcid)
root = ET.fromstring(raw_text)
if root:
abstract_node = root.find("./article/front/article-meta/abstract")
if abstract_node:
abstract_text = ET.tostring(abstract_node, encoding='utf-8', method='text').decode('utf-8')
docs.append({'abstract': abstract_text})
def getAbstracts():
if os.path.exists('./lda/abstracts.json'):
global docs
with open('./lda/abstracts.json', 'r') as f:
docs = json.load(f)
else:
using_dir = CONFIG.JOURNAL_DIRS
filesInDir = []
for directory in using_dir:
filesInDir += [s for s in os.listdir(directory)]
pmcids = []
for f in filesInDir:
pmc_regex = re.search('[\d]+', f)
if pmc_regex:
pmc = pmc_regex.group(0)
pmcids.append(pmc)
threads = []
numThreads = 16
numEntriesPerThread = int(len(pmcids)/numThreads)
remainder = len(pmcids)%numThreads
startIdx = 0
endIdx = numEntriesPerThread
for i in range(numThreads):
if i < remainder:
endIdx+=1
print(startIdx, endIdx, endIdx-startIdx)
pmcid_arr = pmcids[startIdx:endIdx]
t = pubThread(i, pmcid_arr)
threads.append(t)
t.start()
startIdx = endIdx
endIdx+=numEntriesPerThread
for t in threads:
t.join()
with open('./abstracts.json', 'w') as f:
json.dump(docs, f)
def getDictionary(getMatrix=False):
global dictionary
doc_clean = []
if getMatrix:
getAbstracts()
doc_clean = [clean(doc['abstract']).split() for doc in docs]
if os.path.exists('./lda/abstract.dict'):
# with open('./abstract.dict', 'r') as f:
dictionary = corpora.Dictionary.load('./lda/abstract.dict')
else:
dictionary = corpora.Dictionary(doc_clean)
dictionary.save('./lda/abstract.dict')
return [dictionary.doc2bow(doc) for doc in doc_clean]
def generateLDAModel():
doc_term_matrix = getDictionary(getMatrix=True)
Lda = gensim.models.ldamodel.LdaModel
ldamodel = Lda(doc_term_matrix, num_topics=20, id2word=dictionary, alpha='auto', passes=50)
ldamodel.save('./lda/lda.model')
return ldamodel
def main():
ldamodel = None
if os.path.exists('./lda/lda.model'):
ldamodel = gensim.models.ldamodel.LdaModel.load('./lda/lda.model')
else:
ldamodel = generateLDAModel()
data = 'protein protein interaction'
getDictionary()
word_vec = dictionary.doc2bow(data.split())
a = list(sorted(ldamodel[word_vec], key=lambda x: x[1]))
print(ldamodel.print_topic(a[-1][0], topn=20), a[-1])
print(ldamodel.print_topic(a[-2][0], topn=20), a[-2])
# topics = ldamodel.print_topics(num_topics=20, num_words=10)
# for topic in topics:
# print(topic)
# print()
if __name__ == '__main__':
main()