Anime and the People

Inspiration

As an avid fan and watcher of anime, as well as a user of the MyAnimeList (MAL) to track my own watching, I am personally interested in this topic. Knowing answers to these questions could help provide insights and understand the interests and anime watching habits of users, while also possibly providing them a better watching experience.

What it does

Using data scraped from MyAnimeList.net, creates visualizations on genre statistics, and also creates a recommender that which recommends similar shows to a given input.

How we built it

Using datasets on anime entries and user ratings, cleaned our data and manipulated it to calculate genre statistics. For the recommender, I used a nearest neighbor algorithm from sklearn and collaborative filtering to determine the distances between each entry.

Challenges we ran into

A challenge that I ran into was mainly determining what type of model should be used for a recommender system, as well as learning to preprocess that data for use in the machine learning model.

Accomplishments that we're proud of

Successfully completing this project, which was a topic I was personally interested, as well as learning more about recommender systems and machine learning in general.

What we learned

The work process between creating your own project, as well as how to work with large sets of data and, working with with machine learning.

What's next for Anime and the People

Improving upon the recommender, implementing a search engine to allow users to search for an entry and input that, rather than having to copy and paste a name from the myanimelist page. Also, working on the problem with bias towards more popular entries.

Built With

• python