-
Web crawling using BeautifulSoup, used 2 main domains and 100 related articles for each main domain
-
Used namuwiki articles which are written in korean as source
-
Then I made word clouds from collected data
-
Attained similar words using Word2Vec
Note: I wrote this notebook in jupyter notebook, not colab. You would also need a file of shape that you want for wordcloud and a font file in your local environment.
-
Web crawling from dynamic website using selenium
-
Collected reviews and ratings of John Wick4 from imdb
-
Sentiment classification and rating prediction of movie reviews using Gemini & Huggingface (make sure to use colab T4 GPU)
- I used this dataset to recommend movies based on user profile
- I used Chatgpt API and streamlit for this. Give a topic as input then it generates a poetry for you.