Games and big data: A scalable multi-dimensional churn prediction model

Abstract

The emergence of mobile games has caused a paradigm shift in the video-game industry. Game developers now have at their disposal a plethora of information on their players, and thus can take advantage of reliable models that can accurately predict player behavior and scale to huge datasets. Churn prediction, a challenge common to a variety of sectors, is particularly relevant for the mobile game industry, as player retention is crucial for the successful monetization of a game. In this article, we present an approach to predicting game abandon based on survival ensembles. Our method provides accurate predictions on both the level at which each player will leave the game and their accumulated playtime until that moment. Further, it is robust to different data distributions and applicable to a wide range of response variables, while also allowing for efficient parallelization of the algorithm. This makes our model well suited to perform real-time analyses of churners, even for games with millions of daily active users.

Aging Agents: Cross Generational Analysis of Behavior and Need Satisfaction Among Players of Tom Clancy’s The Division 2

“…We also showed that duration of play and playing in groups correlates positively with age. Playtime is among the most popular measures of engagement (Wirth et al 2013) as well as player churn prediction (Bertens et al 2017). More playtime for older adults other than more available free time for the retired individuals, could be connected to the trend of lower perception of Mastery among older generations.…”

Micro- and macro-level churn analysis of large-scale mobile games

“…Some recent research has started to use more advanced models. [10,14,16] propose to use survival [17] achieve good performance by using convolution neural networks (CNN) and long short-term memory networks (LSTM). There are also several recent deep-learning-based studies [33,34,35] for non-game churn prediction problems, which report better performance.…”

Modelling Early User-Game Interactions for Joint Estimation of Survival Time and Churn Probability

“…Additionally, we acquired a single context feature specifying the game context from where the metrics were originated. For determining the targets for our survival and churn estimation tasks, we leveraged existing literature on churn prediction [1]- [4], [19], [22], [23], [26] and survival analysis [6], [20], [21], [26], extending existing rules to accommodate the need to define churn and survival time in single player games with a defined life cycle (i.e. non-GaaS games).…”