DOTA 2 match prediction through deep learning team fight models

Ke, Cheng Hao; Deng, Haozhang; Xu, Congda; Li, Jiong; Gu, Xingyun; Klabjan, Diego; Sifa, Rafet; Drachen, Anders; Demediuk, Simon

doi:10.1109/cog51982.2022.9893647

Cited by 6 publications

(3 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the area of predicting results in MOBA games, specifically in the context of DOTA 2, studies have been conducted that contribute to the advancement of this field. Ke et al [6] propose an innovative approach to identify and define team fights as crucial events during DOTA 2 matches. The main objective of this study is to explore the potential of team fight information in real-time prediction of match results.…”

Section: A Prediction Of Results In Moba Gamesmentioning

confidence: 99%

League of Legends: Real-Time Result Prediction

Junior,

Campelo

2023

Anais Do XVI Congresso Brasileiro De Inteligência Computacional

View full text Add to dashboard Cite

This paper presents a study on the prediction of outcomes in matches of the electronic game League of Legends (LoL) using machine learning techniques. With the aim of exploring the ability to predict real-time results, considering different variables and stages of the match, we highlight the use of unpublished data as a fundamental part of this process. With the increasing popularity of LoL and the emergence of tournaments, betting related to the game has also emerged, making the investigation in this area even more relevant. A variety of models were evaluated and the results were encouraging. A model based on LightGBM showed the best performance, achieving an average accuracy of 81.62% in intermediate stages of the match when the percentage of elapsed time was between 60% and 80%. On the other hand, the Logistic Regression and Gradient Boosting models proved to be more effective in early stages of the game, with promising results. This study contributes to the field of machine learning applied to electronic games, providing valuable insights into realtime prediction in League of Legends. The results obtained may be relevant for both players seeking to improve their strategies and the betting industry related to the game.

show abstract

Section: A Prediction Of Results In Moba Gamesmentioning

confidence: 99%

League of Legends: Real-Time Result Prediction

Junior,

Campelo

2023

Anais Do XVI Congresso Brasileiro De Inteligência Computacional

View full text Add to dashboard Cite

show abstract

“…At the team level, Schubert et al ( 2016) described a graph-based method for detecting spatio-temporally bounded team encounters (team fights) in Dota 2, and, early on in esports research, noted the potential for predictive analytics in esports. Ke et al (2022) developed a framework for defining and extracting teamfight definitions in Dota 2. They evaluated whether team fights (key but relatively rare events) held a signal useful for match outcome prediction.…”

Section: In-game Event Predictionmentioning

confidence: 99%

Time to Die 2: Improved in-game death prediction in Dota 2

Ringer¹,

Missaoui²,

Hodge³

et al. 2023

Machine Learning with Applications

View full text Add to dashboard Cite

“…The many video game genres that use teams of players as the main game mechanic offer an opportunity to examine multiple types of collective actions and how they relate to team dynamics. Collective behaviors of teams can be included as additional types of input in algorithms predicting the outcomes of eSports matches (Hodge et al, 2021;Ke et al, 2022;Xenopoulos et al, 2020). As sports betting in eSports has increased, so to has interest in using in-game analytics to predict match outcomes (Lopez-Gonzalez & Griffiths, 2018;Sweeney et al, 2021).…”

Section: Future Directions and Applicationsmentioning

confidence: 99%

Using Collective Metrics to Assess Team Dynamics and Performance in eSports

Bonny

2022

International Journal of Gaming and Computer-Mediated Simulations

View full text Add to dashboard Cite

A challenge posed by virtual teams is monitoring team interactions remotely. Research with field-based soccer teams provides evidence that measures of collective behavior can be used to assess the dynamics of sports teams. Collective behaviors calculated using the spatial characteristics of teammates as they moved across the field have been found to vary by the state of the soccer match, including ball possession and proximity to a goal. The present study examined whether similar effects were observed with collective metrics calculated from players of a car-soccer eSport video game. A set of matches were retrieved and used to calculate collective behavior metrics based on the placement of teammates within a virtual arena. A subset of metrics varied by team location and ball possession, aligning with and extending previous field-based soccer research, and correlated with team performance. This suggests that collective behaviors can be used to assess aspects of team dynamics within virtual environments.

show abstract

DOTA 2 match prediction through deep learning team fight models

Cited by 6 publications

References 9 publications

League of Legends: Real-Time Result Prediction

League of Legends: Real-Time Result Prediction

Time to Die 2: Improved in-game death prediction in Dota 2

Using Collective Metrics to Assess Team Dynamics and Performance in eSports

Contact Info

Product

Resources

About