-
Notifications
You must be signed in to change notification settings - Fork 2
/
Copy pathdata_extraction.py
195 lines (160 loc) · 7.37 KB
/
data_extraction.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
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
# -*- coding: utf-8 -*-
"""
Created on Wed Apr 27 17:46:57 2022
@author: Jorge Vasquez,Sahil Chopra,Colm Rooney
"""
import wikipedia
from SPARQLWrapper import SPARQLWrapper, JSON
import pandas as pd
import wptools
import spacy
import requests
import html2text
from itertools import islice
import nltk
import argparse
def data_extraction(k=100,n=3):
# Load English Model
nlp = spacy.load('en_core_web_sm')
#========= Getting the categories ======================================================================
categories=['Airports','Artists','Astronauts','Building','Astronomical_objects','City','Comics_characters',
'Companies','Foods','Transport','Monuments_and_memorials','Politicians','Sports_teams','Sportspeople','Universities_and_colleges','Written_communication']
#-----DEFINE THE NUMBER OF ARTICLES PER CATEGORY ----------
#k=50
#n=3
#---DEFINING LISTS TO STORE THE DATA-----
articles_data=[]
articles_data_2=[]
articles_data_3=[]
articles_data_4=[]
articles_data_5=[]
articles_data_6=[]
#========= Getting the articles from each category =======================================================
sparql=SPARQLWrapper("http://dbpedia.org/sparql/")
# We define the general SPARQL query in order to search 'k' articles in a 'category'
for category in categories:
try:
query=f"""
PREFIX dcterms:<http://purl.org/dc/terms/>
PREFIX dbc:<http://dbpedia.org/resource/Category:>
SELECT ?label WHERE {{
?label
dcterms:subject/skos:broader*
dbc:{category} .
}}
LIMIT {k}
"""
sparql.setQuery(query)
sparql.setReturnFormat(JSON)
results=sparql.query().convert()
for result in results["results"]["bindings"]:
article_c=[]
#---GET THE ARTICLE NAME
article_link=result["label"]["value"]
article=article_link.replace('http://dbpedia.org/resource/','')
article_c=[category,article]
articles_data.append(article_c)
except:
#print("Error")
articles_wkp_list=wikipedia.search(category,k)
for result_wkp in articles_wkp_list:
article_c=[]
article_c=[category,result_wkp]
articles_data.append(article_c)
#----------- GETTING THE DESCRIPTION OF THE ENTITY ----------------------------
for i in range(0,len(articles_data)):
description=''
entity_description=[]
article_name=articles_data[i][1]
try:
result = requests.get('https://www.wikidata.org/w/api.php',
params={'format': 'json',
'action': 'wbsearchentities',
'search': article_name,
'language': 'en'})
result = result.json()
key_id = result['search'][0]['id']
#print(key_id)
except:
pass
try:
page = wptools.page(wikibase=key_id, silent=True)
page.get_wikidata()
description = page.data['description']
#print(description)
except:
pass
entity_description=[articles_data[i][0],articles_data[i][1],description]
articles_data_2.append(entity_description)
#--------------------GET THE WIKIPEDIA PAGE CONTENT--------------------------------------
for i in range(0,len(articles_data)):
content=''
article_content=[]
page=wptools.page(articles_data[i][1],silent=False)
try:
page.get_query()
except:
pass
try:
content=page.data['extract']
except KeyError:
pass
nlp(content)
article_content=[articles_data[i][0],articles_data[i][1],content]
articles_data_3.append(article_content)
#---------------------GET THE INFOBOX-----------------------------------------------------
for i in range(0,len(articles_data)):
infobox=''
article_infobox=[]
page=wptools.page(articles_data[i][1],silent=False)
try:
page.get_parse()
except:
pass
try:
infobox=page.data['infobox']
except KeyError:
pass
article_infobox=[articles_data_3[i][0],articles_data_3[i][1],articles_data_3[i][2],infobox]
articles_data_4.append(article_infobox)
#---------------------GET WIKIDATA STATEMENTS-----------------------------------------------
for i in range(0,len(articles_data)):
statements=''
article_statement=[]
page=wptools.page(articles_data[i][1],silent=False)
try:
page.get_wikidata()
except:
pass
#try:
statements=page.data['wikidata']
#except KeyError:
# pass
article_statement=[articles_data_4[i][0],articles_data_4[i][1],articles_data_4[i][2],articles_data_4[i][3],statements]
articles_data_5.append(article_statement)
#--------------------GETTING ALL THE DATA IN A LIST ---------------------------------
for i in range(0,len(articles_data)):
data=[]
data=[articles_data_5[i][0],articles_data_5[i][1],articles_data_2[i][2],articles_data_5[i][2],articles_data_5[i][3],articles_data_5[i][4]]
articles_data_6.append(data)
#--------------------SAVE ALL THE DATA INTO A DATAFRAME AND TAKE THE ARTICLES WITH MORE THAN n SENTENCES -----------------------------
articles_csv=pd.DataFrame(articles_data_6,columns=['category','article','description','page_content','infobox','statements'])
articles_csv_filter=articles_csv[articles_csv['page_content'].apply(lambda x : sum(1 for dummy in nlp(x).sents)) > n]
#------------------CLEANING THE DATA AND GETTING THE SAME NUMBER OF SENTENCES ------------------------------------------------
articles_csv_filter['page_content']=articles_csv_filter['page_content'].apply(html2text.html2text)
#----We define a funcion in order to get the same number of sentences for each article content
def same_sentences(content_text):
sentences=[sent.strip() for sent in islice(nltk.sent_tokenize(content_text), n)]
return ' '.join(sentences)
articles_csv_filter['page_content']=articles_csv_filter['page_content'].apply(same_sentences)
#--------------------SAVE ALL THE DATA INTO A CSV FILE ------------------------------
articles_csv_filter.to_csv('extracted_data.csv', index=False, header=['category','article','description','page_content','infobox','statements'])
return True
def main(k:int,n:int):
data_extraction(k,n)
if __name__ == "__main__":
parser=argparse.ArgumentParser(description="Data Extraction")
parser.add_argument("--num_articles", type=int, default=50, help="number of articles to extract from each category")
parser.add_argument("--num_sentences", type=int, default=3, help="number of sentences that should be in the article's content")
args = parser.parse_args()
main(args.num_articles, args.num_sentences)