Как находил близкие слова из статьи - word2vec (пакет gensim, anaconda)

import bs4 as bs
import urllib.request
import nltk
from nltk.corpus import stopwords
from gensim.models import Word2Vec
import re
 
scrapped_data = urllib.request.urlopen('https://en.wikipedia.org/wiki/Artificial_intelligence')
article = scrapped_data .read()
 
parsed_article = bs.BeautifulSoup(article,'lxml')
 
paragraphs = parsed_article.find_all('p')
 
article_text = ""
 
for p in paragraphs:
    article_text += p.text
 
# Cleaing the text
processed_article = article_text.lower()
processed_article = re.sub('[^a-zA-Z]', ' ', processed_article )
processed_article = re.sub(r'\s+', ' ', processed_article)
 
# Preparing the dataset
all_sentences = nltk.sent_tokenize(processed_article)
 
all_words = [nltk.word_tokenize(sent) for sent in all_sentences]
 
# Removing Stop Words
 
len_all_words=len(all_words)
for i in range(len_all_words):
    all_words[i] = [w for w in all_words[i] if w not in stopwords.words('english')]   
 
word2vec = Word2Vec(all_words, min_count=2)
 
sim_words = word2vec.wv.most_similar('intelligence')
 
print('similar:\n',sim_words)