Skip to content

Latest commit

 

History

History
31 lines (24 loc) · 1.44 KB

README.md

File metadata and controls

31 lines (24 loc) · 1.44 KB
Raw

MyAnimeList Recommender System

This project was initially built for the Udacity Machine Learning Engineer Nanodegree.

Project Status: Active

Project Intro/Objective

The purpose of this project is to develop the best suitable collaborative filtering user-based recommender system for the MyAnimeList portal. MyAnimeList is currently the most relevant portal for Anime and Manga content and today it counts with over 4M users interacting and scoring the available shows.

Methods Used

  • Collaborative filtering application (Recommendation system)

Technology needed for the project to work

  • Python 3.6
  • Notebook Jupyter
  • scikit-surprise 1.0.6
  • Pandas 0.24.2
  • Numpy 1.16.2
  • Scipy 1.21
  • Matplotlib 3.0.3
  • The datasets used in the project are all available in Kaggle

Project Content

As from today, this project relies on

  • Do some exploration so the profile of the MyAnimeList users could be understood
  • Make all the data cleaning and wrangling needed so the collaborative filtering algorithms could be tested
  • Train and test all the available algorithms at the chosen lib and choose the best performing one
  • Create an easy way of checking the recommendations for each user and even develop a MVP of an item-based recommender

Contact