**By using web scrapping method, extracted all words from any web page.
**Text cleaning with nltk('stopwords')
**With Glove and Spacy trained models, word embedding.
**Finally, Cosine Similarity provided us to find the most relevant words from web page.