Anime recommendation system based on Surprise - a Python scikit for recommender systems - made interactive through tkinter-based CustomTkinter library.
Clone the repository
git clone https://github.com/beltran99/AnimeRecommender.gitThis RecSys is trained on Kaggle's Anime Recommendations Database, so the next step consists on downloading it. You can easily download the dataset from your preferred web browser and unzip it on the data folder.
However, you can also download it via command-line through Kaggle’s public API implemented in Python:
- Install the kaggle cli tool and authenticate yourself as indicated in this tutorial.
pip install kaggle- You can check that the tool works correctly by searching for datasets containing the word "anime".
kaggle datasets list -s "anime"- Create the data subfolder
mkdir data/external
cd data/external- Download the dataset
kaggle datasets download -d CooperUnion/anime-recommendations-database- Unzip the downloaded dataset
unzip anime-recommendations-database.zip- Search for animes
- Rate the animes you searched for
- Receive recommendations based on your rating history
You can invoke the app with the followning command:
python3 -m gui.appThen you can use the GUI to search for your favorite animes with the top search bar and add your ratings with the corresponding menus for each of the animes. Once you have rated enough animes, you will be able to receive a recommendation and a button for that purpose will pop up.