Project: JOGEN
This is a project of the CIIS course (WS 2021/22). The content below was created by Nikita Agrawal, Nimisha Vernekar, Rayan Hamid Mohiuddin.
Motivation & Idea:
This is for all the Anime lovers out there. Do you ever wonder how helplful it would be to maintain the list of all the Anime that you are watching with just a few click of your fingers? Would you like to find more Anime that you might like? Then you have come to the right place.
JOGEN helps you to maintain a list of all the Anime you have watched, or are watching, or also would like to watch in the future. It caters Anime recommendations
personalised for you.
Moreover the app provides two different kind of recommendations:
• Standard Recommendations – recommendations depending upon what you have liked
• Advanced Recommendations – recommendations on the basis of Artstyle, Sound track, Humour and Vibes.
App and its working:
• User must login using their “MyAnimeList.net” account into “Jogen”.
• Once user login he can see the list of Anime he is currently watching.
• User can navigate through different lists using the navigation panel.
• User can opt for basic recommendations and get recommendations based on other users who also have similar taste in animes as him/her.
• User can select advanced recommendations and get recommendations specific to Art Style, Sound track, Humour and Vibes similar to what he/she likes.
User must have an “MyAnimeList.net” account and must have at least one anime added to the account for recommendations.
For Basic Recommendations, we take in very similar anime based on scores/ratings.
For Advanced Recommendations, we parse the “MyAnimeList.net” website for user reviews. Our code uses the Beautiful Soup Python package for web scraping and maintain separate datasets for each category in which the data is inserted by specifying certain keywords for each category, namely, Art style, Sound Track, Humour and Vibes. For example, let us take a review : “The art fits so well with the grandiose nature of everything going on in the show”, this will be categorized into the art style dataset and so on for the other categories as well.