def viewSearchbyTitle():
if request.method == 'POST':
mystring = "Title"
query = request.form['namesearchbytitle']
data = pickle.load(open(r"DataBase\data_file.pkl", "rb"))
titles = pickle.load(open(r"DataBase\title_file.pkl", "rb"))
auto_tag = pickle.load(open(r"DataBase\svos_file.pkl", "rb"))
summary = pickle.load(open(r"DataBase\summary_file.pkl", "rb"))
corpus = pickle.load(open(r"DataBase\title_corpus.pkl", "rb"))
bm25 = search_by_BM25(corpus)
tokenized_query, old_query, new_query = clean_query(query.lower())
indexes, results = bm25.get_top_n(tokenized_query, data, n=5)
results_titles = []
results_summaries = []
results_tags = []
for i in indexes:
results_titles.append(titles[i])
results_summaries.append(summary[i])
if auto_tag[i] != []:
results_tags.append(list(set(random.choices(auto_tag[i],
k=3))))
else:
results_tags.append(['No Auto tags'])
text = []
for i in results:
text_to_show = " ".join(sent_tokenize(i)[:2])
if text_to_show != '':
text.append(text_to_show + '....')
else:
text.append(i)
# text = results
title = results_titles
summaries = results_summaries
tags = results_tags
title_len = len(title)
document_file = pickle.load(open(r"DataBase\document_file.pkl", "rb"))
extension_list = []
for i in indexes:
extension_list.append(document_file[i]["extension"])
# return render_template('searchbyTitle.html', text=text, tag=query, title=title, summaries=summaries, tags=tags, title_len = title_len)
return render_template('searchbyText.html',
text=text,
tag=query,
title=title,
summaries=summaries,
tags=tags,
type=mystring,
title_len=title_len,
old_query=old_query,
new_query=new_query,
extension_list=extension_list)
def viewSearchbyTitle(the_text):
mystring = "Title"
query = the_text
corpus = data_for_title
bm25 = search_by_BM25(corpus)
tokenized_query, old_query, new_query = clean_query(query.lower())
indexes, results = bm25.get_top_n(tokenized_query, data, n=5)
results_titles = []
results_summaries = []
results_tags = []
for i in indexes:
results_titles.append(titles[i])
results_summaries.append(summary[i])
if auto_tag[i] != []:
results_tags.append(list(set(random.choices(auto_tag[i], k=3))))
else:
results_tags.append(['No Auto tags'])
text = []
for i in results:
text_to_show = " ".join(sent_tokenize(i)[:2])
if text_to_show != '':
text.append(text_to_show + '....')
else:
text.append(i)
# text = results
title = results_titles
summaries = results_summaries
tags = results_tags
title_len = len(title)
extension_list = []
for i in indexes:
extension_list.append(document_file[i]["extension"])
# return render_template('searchbyTitle.html', text=text, tag=query, title=title, summaries=summaries, tags=tags, title_len = title_len)
return render_template('searchbyText.html',
text=text,
tag=query,
title=title,
summaries=summaries,
tags=tags,
type=mystring,
title_len=title_len,
old_query=old_query,
new_query=new_query,
extension_list=extension_list)
def viewSearchbyText():
if request.method == 'POST':
mystring = "Text"
query = request.form['namesearchbytext']
# query = "Corona Virus"
data = pickle.load(open(r"DataBase\data_file.pkl", "rb"))
titles = pickle.load(open(r"DataBase\title_file.pkl", "rb"))
corpus = pickle.load(open(r"DataBase\wiki_corpus_file.pkl", "rb"))
bm25 = search_by_BM25(corpus)
tokenized_query, old_query, new_query = clean_query(query.lower())
indexes, results = bm25.get_top_n(tokenized_query, data, n=5)
results_titles = []
for i in indexes:
results_titles.append(titles[i])
text = []
for i in results:
text_to_show = " ".join(sent_tokenize(i)[:2])
if text_to_show != '':
text.append(text_to_show + '....')
else:
text.append(i)
title = results_titles
title_len = len(title)
document_file = pickle.load(open(r"DataBase\document_file.pkl", "rb"))
extension_list = []
for i in indexes:
if i < 10909:
extension_list.append('wikipedia')
else:
extension_list.append(document_file[i - 10909]["extension"])
return render_template('searchbyText.html',
text=text,
tag=query,
title=title,
type=mystring,
title_len=title_len,
old_query=old_query,
new_query=new_query,
extension_list=extension_list)
return lemmed
if __name__ == '__main__':
data = pickle.load(open(r"DataBase/data_file.pkl", "rb"))
titles = pickle.load(open(r"DataBase/title_file.pkl", "rb"))
option = input("Enter option of search")
if option == 'default search':
corpus = pickle.load(open(r"DataBase/corpus_file.pkl", "rb"))
elif option == 'tag search':
corpus = pickle.load(open(r"DataBase/tags_pickle.pkl", "rb"))
elif option == 'title search':
corpus = pickle.load(open(r"DataBase/title_corpus.pkl", "rb"))
else:
print('Not valid option')
bm25 = search_by_BM25(corpus)
query = input("Enter query")
tokenized_query = clean_query(query.lower())
indexes, results = bm25.get_top_n(tokenized_query, data, n=20)
results_titles = []
for i in indexes:
results_titles.append(titles[i])
for i in range(10):
print(f"Title_{i}: {results_titles[i]}")
print(f"\nText_{i}: {results[i]}")
print('\n\n')
