Anime-Finder

An anime recommendation system. It was developed using Anime Ratings Database from which was created using data from .

Requirements

1. Python 3
2. scikit-learn
3. scipy
4. numpy
5. pandas

Setup

1. Download and extract the dataset to the same directory as the project.
2. Run compute_svd.py as python3 compute_svd.py to build an Matrix Decomposition based Anime Similarity model and compute the 10 most similar anime for each anime.
3. Run create_anime_json.py to convert anime.csv to JSON format such that the anime objects are sorted by rating in decreasing order.
4. Run create_similar_json.py to convert the similar anime list to a format containing more detail about each anime.

Execution

Start an HTTP server in the project directory. A simple way to do so is to run python3 -m http.server and to view the app, open http://localhost:8000.