Skip to content

Includes data analysis from matches pulled from kaggle and pca clustering of all champions.

Notifications You must be signed in to change notification settings

peanutshawny/league-of-legends-pca

Repository files navigation

league-of-legends

Includes pca/t-sne clustering of all champion images taken from the Riot API.

-- Project Status: [Inactive]

Project Intro/Objective

Methods Used

  • PCA
  • t-SNE
  • kmeans
  • Image pre-processing

Technologies/Languages

  • Python & Jupyter Notebook
  • Riot Games Developer API
  • Cassiopeia (Python API wrapper for Riot API)

Project Description

League of Legends is a team oriented video game where on two team teams (with 5 players in each) compete for objectives and kills. Being an avid League of Legends player for over 8 years, I thought it was about time that I took some data from the game and made an effort to take a deeper dive into some of the champions.

I wanted to cluster champions by both their in-game stats and their appearance. This will hopefully yield some insights about how champions were designed and also some cool visualizations.

Description of my Current Process

Check out my notebooks and scripts for more details!

Conclusion & Next Steps

Good project all in all! I learned a lot about different dimensionality reduction techniques in addition to gaining some skills in image filter and working with apis.

Next Steps:

  • Perform PCA/t-SNE on champion splash art as that would theoretically provide more data to draw features from
  • Look for matches and perform some exploratory analysis
  • Further modularize code, including building different python files for visualization

Contact

Feel free to email me at [email protected] if you want to talk about this project or data science in general!

About

Includes data analysis from matches pulled from kaggle and pca clustering of all champions.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published