data

Data processing

This folder contains the data processing script for the Anime Data Visualization. Each script is a Python Jupyter notebook and will output a JSON file once ran, which can be manually tweaked for formatting. Final JSON files used by the data visualization app are placed in absolute folder /app/public/data.

Prerequisites

To run the scripts you will need:

• Python 3 (link): scripting and programming runtime and environment
• Jupyter Notebook (link): web-based development application for interactive Python scripts
• Pandas (link): data analysis library for Python
• JikanPy (link): Python wrapper library around the Jikan API
• The original dataset CSVs which can be found here, placed in relative folder ./raw (you may need to create it)

Each .ipynb can be opened in your local Jupyter Notebook instance as a standalone script.

Scripts description

• Genres.ipynb: takes as input raw/anime_cleaned.csv and produces genre_data.json and genre_top_animes_data.json, containing the data for the genres bubble diagram.

• History.ipynb: takes as input raw/AnimeList.csv and produces history.json, containing the data for the histogram

• Actors.ipynb: takes as input raw/anime_cleaned.csv and fetches data from the Jikan API. It outputs vA_datasets.json and vA_infos.json (the latter is to be placed in /app/src/pages/chord for bundling) which contain the data for the actors chord diagram.

• Studios.ipynb: takes as input raw/anime_cleaned.csv and produces sankey_dataset.json which contains the data for the studios sankey diagram.