Loading…

Sam Kenkel

Data Science, Machine Learning, DevOps, CCNA, ACSR
Learn More

Lol_Scout 3: Final Modelling and Results

Background:Summary of the previous posts.

In the 5 v 5 Videogame/Esport two teams of 5 players compete against each other. I have gathered Data using the API from riot games. I’m trying to use machine learning to predict wins or losses based on the characters (Champions) that players choose, and those player’s skill/ practice with those champions.

This is the 3rd of 3 blog posts about my process and discoveries working with data from Riot’s online game, League of Legends. ¬†The code I wrote for initial sanity check modelling work can be found¬†here. The feature engineering/Data Prep code is here. The code for my ‘final’ models is here

I have scraped data using Riot’s Api. I have tested an initial model that involved only which champions were chosen using a latent factors neural net with Keras, and Xgboost (Boosted decission trees), and I have tried to add to this data the metrics I have gathered, however insufficient they are, about player experience with specific characters(champions) and tried several method of imputation for the (large amount of) missing data.

Below are my experiments which show that this additional data is reducing my model’s accuracy.