A Player Behavior Model for Predicting Win-Loss Outcome in MOBA Games

Lan, Xuan; Duan, Lei; Chen, Wen; Qin, Ruiqi; Nummenmaa, Timo; Nummenmaa, Jyrki

doi:10.1007/978-3-030-05090-0_41

Cited by 18 publications

(7 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Another of the objectives found in selected researches is the prediction of the winning team in a game. This category uses variables that are known at the beginning of the game [15] [27] [60] [43] [31] [65] [24] [25], at any time during the game [14] [62] or variables known at the end of a game in order to understand their influence on the result [66] [54] [20]. Therefore, this category will be called victory.…”

Section: Victorymentioning

confidence: 99%

Relevant Independent Variables on MOBA Video Games to Train Machine Learning Algorithms

López-Guzmán¹,

Bustacara-Medina²

2021

CSRN

View full text Add to dashboard Cite

Popularity of Multiplayer Online Battle Arena (MOBA) video games has grown considerably, its popularity as well as the complexity of their playability, have attracted the attention in recent years of researchers from various areas of knowledge and in particular how they have resorted to different machine learning techniques. The papers reviewed mainly look for patterns in multidimensional data sets. Furthermore, these previous researches do not present a way to select the independent variables (predictors) to train the models. For this reason, this paper proposes a list of variables based on the techniques used and the objectives of the research. It allows to provide a set of variables to find patterns applied in MOBA videogames. In order to get the mentioned list, the consulted works were grouped by the used machine learning techniques, ranging from rule-based systems to complex neural network architectures. Also, a grouping technique is applied based on the objective of each research proposed.

show abstract

Section: Victorymentioning

confidence: 99%

Relevant Independent Variables on MOBA Video Games to Train Machine Learning Algorithms

López-Guzmán¹,

Bustacara-Medina²

2021

CSRN

View full text Add to dashboard Cite

show abstract

“…champions) lineups in Dota 2. [14] used neural networks to extract player behavioral patterns to predict match outcomes. These methods do not consider at the same time the individual variability of users and the characteristics of champions they select, therefore lacking contextual awareness.…”

Section: Related Workmentioning

confidence: 99%

Individualized Context-Aware Tensor Factorization for Online Games Predictions

Jiang,

Lerman,

Ferrara

2021

Preprint

View full text Add to dashboard Cite

Individual behavior and decisions are substantially influenced by their contexts, such as location, environment, and time. Changes along these dimensions can be readily observed in Multiplayer Online Battle Arena games (MOBA), where players face different in-game settings for each match and are subject to frequent game patches. Existing methods utilizing contextual information generalize the effect of a context over the entire population, but contextual information tailored to each individual can be more effective. To achieve this, we present the Neural Individualized Context-aware Embeddings (NICE) model for predicting user performance and game outcomes. Our proposed method identifies individual behavioral differences in different contexts by learning latent representations of users and contexts through non-negative tensor factorization. Using a dataset from the MOBA game League of Legends, we demonstrate that our model substantially improves the prediction of winning outcome, individual user performance, and user engagement.

show abstract

“…To the best of our knowledge, our work is the first to succeed in employing players' immediate history to improve a win prediction model; previous work by Grutzik et al [2017], on Esports win prediction in DotA 2, made an attempt at this, using neural networks and rolling statistics for the last 10 professional matches. Other work has used hierarchical attentionbased networks to recommend purchasable items in the mobile MOBA King of Glory , and there is much research on recurrent models for in-game prediction [Lan et al 2018]. In terms of in-production recommendation and coaching tools for League, the most popular is Blitz, with over 1.5 million users, providing matchup-based champion suggestions, optimal pre-match runes, item build paths, and informative post-game analysis.…”

Section: Related Workmentioning

confidence: 99%

Scalable Psychological Momentum Forecasting in Esports

White,

Romano

2020

Preprint

View full text Add to dashboard Cite

The world of competitive Esports and video gaming has seen and continues to experience steady growth in popularity and complexity. Correspondingly, more research on the topic is being published, ranging from social network analyses to the benchmarking of advanced artificial intelligence systems in playing against humans. In this paper, we present ongoing work on an intelligent agent recommendation engine that suggests actions to players in order to maximise success and enjoyment, both in the space of in-game choices, as well as decisions made around play session timing in the broader context. By leveraging temporal data and appropriate models, we show that a learned representation of player psychological momentum, and of tilt, can be used, in combination with player expertise, to achieve state-of-the-art performance in preand post-draft win prediction. Our progress toward fulfilling the potential for deriving optimal recommendations is documented.

show abstract

A Player Behavior Model for Predicting Win-Loss Outcome in MOBA Games

Cited by 18 publications

References 12 publications

Relevant Independent Variables on MOBA Video Games to Train Machine Learning Algorithms

Relevant Independent Variables on MOBA Video Games to Train Machine Learning Algorithms

Individualized Context-Aware Tensor Factorization for Online Games Predictions

Scalable Psychological Momentum Forecasting in Esports

Contact Info

Product

Resources

About