Can I read Predicting DotA2 Match Winners Using Regression on EtoBox?
Predicting DotA2 Match Winners Using Regression by Jair Polanco is a document available to read on EtoBox.
What is Predicting DotA2 Match Winners Using Regression about?
This document summarizes a research project that aims to predict the winning side in Dota 2 matches using logistic regression models. The researchers collected match and player data using APIs and filtered out invalid matches. They tested a baseline model that only considered hero lineups, representing each hero as a feature. This baseline model achieved a training error of 28% and testing error of 45%, indicating hero lineups alone were not enough to accurately predict matches. The researchers hypothesized
- Author
- Jair Polanco
- Language
- EN