Anime Recommendation and Rating App

Inspiration

Using the old anime search engines of the past plagued every anime lover. And these recommendations always seemed to be inconsistent, with many of them not matching the user's interests. The websites are terrible to use, and they just rotate the same anime again and again, which might be good for new anime watchers, but not sufficient for actual enthusiasts.

What it does

The Anime Recommendation and Rating App does exactly what it suggests. A user is allowed to rate random anime on a scale of one to five, with the option of expressing that he or she has not watched the anime. And then, the user can also look at recommendations to watch new anime, with three exciting categories. The Creator's Picks, which are anime hand picked by us that have weathered the test of time. Your Recommendations are anime that are tailored for you. Anime of a particular genre that are best suited for your tastes, that you can try out. And finally, the Busts. Anime that most people have found quirky, or "bad", but have the possibility to be a diamond in the rough for the right viewers. The admin also has the power to add anime to the database, and keep the database up to date with the latest anime.

How we built it

For the database, we used a list of all of the anime, and uploaded it to the database. We do have a working cockroachdb solution, which can be implemented once cockroach db is up and running. The database can also be locally run, with a python script that uses sql alchemy to communicate with the database. The website is built on next.js, and it connects to the python backend using an API.

Challenges we ran into

Uploading the csv file into cockroach db was slightly difficult, but eventually, we were able to use a python script that was able to do that.

Accomplishments that we're proud of

Mutating data on the client side using swr Using postgres to perform recommendation searches

What we learned

Learning next.js to create a website Using cockroach db to deploy a database Connecting the website and python back end using APIs

What's next for Anime Recommendation and Rating App

There are a lot of features that could still be implemented and are along the pipeline. A searching feature could be added for admins and users. It is half completed, and is close to implementation. The recommendation algorithm could be improved significantly with machine learning. Multiple client features, like stacking genres for sorting, and creating anime watch lists could be added.

Built With

• flask
• next.js
• postgresql
• python
• railway
• sqlalchemy