TY - GEN
T1 - Retention Prediction in Sandbox Games with Bipartite Tensor Factorization
AU - Sifa, Rafet
AU - Fedell, Michael
AU - Franklin, Nathan
AU - Klabjan, Diego
AU - Ram, Shiva
AU - Venugopal, Arpan
AU - Demediuk, Simon
AU - Drachen, Anders
N1 - Funding Information:
Part of this work was jointly funded by the Audience of the Future programme by UK Research and Innovation through the Industrial Strategy Challenge Fund (grant no.104775) and supported by the Digital Creativity Labs (digitalcreativity.ac.uk), a jointly funded project by EPSRC/AHRC/ Innovate UK under grant no. EP/M023265/1. Additionally, part of this research was funded by the Federal Ministry of Education and Research of Germany as part of the competence center for machine learning ML2R (01—S18038A).
Publisher Copyright:
© 2020, Springer Nature Switzerland AG.
PY - 2020
Y1 - 2020
N2 - Open world video games are designed to offer free-roaming virtual environments and agency to the players, providing a substantial degree of freedom to play the games in the way the individual player prefers. Open world games are typically either persistent, or for single-player versions semi-persistent, meaning that they can be played for long periods of time and generate substantial volumes and variety of user telemetry. Combined, these factors can make it challenging to develop insights about player behavior to inform design and live operations in open world games. Predicting the behavior of players is an important analytical tool for understanding how a game is being played and understand why players depart (churn). In this paper, we discuss a novel method of learning compressed temporal and behavioral features to predict players that are likely to churn or to continue engaging with the game. We have adopted the Relaxed Tensor Dual DEDICOM (RTDD) algorithm for bipartite tensor factorization of temporal and behavioral data, allowing for automatic representation learning and dimensionality reduction.
AB - Open world video games are designed to offer free-roaming virtual environments and agency to the players, providing a substantial degree of freedom to play the games in the way the individual player prefers. Open world games are typically either persistent, or for single-player versions semi-persistent, meaning that they can be played for long periods of time and generate substantial volumes and variety of user telemetry. Combined, these factors can make it challenging to develop insights about player behavior to inform design and live operations in open world games. Predicting the behavior of players is an important analytical tool for understanding how a game is being played and understand why players depart (churn). In this paper, we discuss a novel method of learning compressed temporal and behavioral features to predict players that are likely to churn or to continue engaging with the game. We have adopted the Relaxed Tensor Dual DEDICOM (RTDD) algorithm for bipartite tensor factorization of temporal and behavioral data, allowing for automatic representation learning and dimensionality reduction.
KW - Behavioral analytics
KW - Business intelligence
KW - Tensor factorization
UR - http://www.scopus.com/inward/record.url?scp=85088520279&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85088520279&partnerID=8YFLogxK
U2 - 10.1007/978-3-030-52249-0_21
DO - 10.1007/978-3-030-52249-0_21
M3 - Conference contribution
AN - SCOPUS:85088520279
SN - 9783030522483
T3 - Advances in Intelligent Systems and Computing
SP - 297
EP - 308
BT - Intelligent Computing - Proceedings of the 2020 Computing Conference
A2 - Arai, Kohei
A2 - Kapoor, Supriya
A2 - Bhatia, Rahul
PB - Springer
T2 - Science and Information Conference, SAI 2020
Y2 - 16 July 2020 through 17 July 2020
ER -