Combining Sequential and Aggregated Data for Churn Prediction inCasual Freemium Games

Jeppe Theiss Kristensen; Paolo Burelli

doi:10.1109/CIG.2019.8848106

Combining Sequential and Aggregated Data for Churn Prediction inCasual Freemium Games

Jeppe Theiss Kristensen, Paolo Burelli

Research output: Conference Article in Proceeding or Book/Report chapter › Article in proceedings › Research › peer-review

Abstract

In freemium games, the revenue from a player comes from the in-app purchases made and the advertisement to which that player is exposed. The longer a player is playing the game, the higher will be the chances that he or she will generate a revenue within the game.
Within this scenario, it is extremely important to be able to detect promptly when a player is about to quit playing (churn) in order to react and attempt to retain the player within the game, thus prolonging his or her game lifetime.
In this article we investigate how to improve the current state-of-the-art in churn prediction by combining sequential and aggregate data using different neural network architectures.
The results of the comparative analysis show that the combination of the two data types grants an improvement in the prediction accuracy over predictors based on either purely sequential or purely aggregated data.

Original language	English
Title of host publication	Proceedings of the IEEE Conference on Games
Number of pages	8
Publisher	IEEE
Publication date	30 Jul 2019
ISBN (Print)	978-1-7281-1885-7
ISBN (Electronic)	978-1-7281-1884-0
DOIs	https://doi.org/10.1109/CIG.2019.8848106
Publication status	Published - 30 Jul 2019
Event	IEEE Conference on Games (2019) - Queen Mary University of London, London, United Kingdom Duration: 20 Aug 2019 → 23 Aug 2019 http://www.ieee-cog.org

Conference

Conference	IEEE Conference on Games (2019)
Location	Queen Mary University of London
Country/Territory	United Kingdom
City	London
Period	20/08/2019 → 23/08/2019
Internet address	http://www.ieee-cog.org

Keywords

Freemium games
Churn prediction
Neural network architectures
Sequential data
Aggregate data

Access to Document

10.1109/CIG.2019.8848106

mainAccepted author manuscript, 516 KB

ALGO: Autonomous Live Game Operations
Kristensen, J. T. (PI) & Burelli, P. (PI)
IFD - Innovation Fund Denmark
01/04/2019 → 30/07/2022
Project: Research

Cite this

@inproceedings{53485f37441b42c4aadb3bed8975b104,

title = "Combining Sequential and Aggregated Data for Churn Prediction inCasual Freemium Games",

abstract = "In freemium games, the revenue from a player comes from the in-app purchases made and the advertisement to which that player is exposed. The longer a player is playing the game, the higher will be the chances that he or she will generate a revenue within the game.Within this scenario, it is extremely important to be able to detect promptly when a player is about to quit playing (churn) in order to react and attempt to retain the player within the game, thus prolonging his or her game lifetime.In this article we investigate how to improve the current state-of-the-art in churn prediction by combining sequential and aggregate data using different neural network architectures.The results of the comparative analysis show that the combination of the two data types grants an improvement in the prediction accuracy over predictors based on either purely sequential or purely aggregated data.",

keywords = "Freemium games, , Churn prediction, , Neural network architectures, , Sequential data, , Aggregate data, Freemium games, , Churn prediction, , Neural network architectures, , Sequential data, , Aggregate data",

author = "Kristensen, {Jeppe Theiss} and Paolo Burelli",

year = "2019",

month = jul,

day = "30",

doi = "10.1109/CIG.2019.8848106",

language = "English",

isbn = "978-1-7281-1885-7",

booktitle = "Proceedings of the IEEE Conference on Games",

publisher = "IEEE",

address = "United States",

note = "IEEE Conference on Games (2019), IEEE COG ; Conference date: 20-08-2019 Through 23-08-2019",

url = "http://www.ieee-cog.org",

}

TY - GEN

T1 - Combining Sequential and Aggregated Data for Churn Prediction inCasual Freemium Games

AU - Kristensen, Jeppe Theiss

AU - Burelli, Paolo

PY - 2019/7/30

Y1 - 2019/7/30

N2 - In freemium games, the revenue from a player comes from the in-app purchases made and the advertisement to which that player is exposed. The longer a player is playing the game, the higher will be the chances that he or she will generate a revenue within the game.Within this scenario, it is extremely important to be able to detect promptly when a player is about to quit playing (churn) in order to react and attempt to retain the player within the game, thus prolonging his or her game lifetime.In this article we investigate how to improve the current state-of-the-art in churn prediction by combining sequential and aggregate data using different neural network architectures.The results of the comparative analysis show that the combination of the two data types grants an improvement in the prediction accuracy over predictors based on either purely sequential or purely aggregated data.

AB - In freemium games, the revenue from a player comes from the in-app purchases made and the advertisement to which that player is exposed. The longer a player is playing the game, the higher will be the chances that he or she will generate a revenue within the game.Within this scenario, it is extremely important to be able to detect promptly when a player is about to quit playing (churn) in order to react and attempt to retain the player within the game, thus prolonging his or her game lifetime.In this article we investigate how to improve the current state-of-the-art in churn prediction by combining sequential and aggregate data using different neural network architectures.The results of the comparative analysis show that the combination of the two data types grants an improvement in the prediction accuracy over predictors based on either purely sequential or purely aggregated data.

KW - Freemium games

KW - Churn prediction

KW - Neural network architectures

KW - Sequential data

KW - Aggregate data

KW - Freemium games

KW - Churn prediction

KW - Neural network architectures

KW - Sequential data

KW - Aggregate data

U2 - 10.1109/CIG.2019.8848106

DO - 10.1109/CIG.2019.8848106

M3 - Article in proceedings

SN - 978-1-7281-1885-7

BT - Proceedings of the IEEE Conference on Games

PB - IEEE

T2 - IEEE Conference on Games (2019)

Y2 - 20 August 2019 through 23 August 2019

ER -

Combining Sequential and Aggregated Data for Churn Prediction inCasual Freemium Games

Abstract

Conference

Keywords

Access to Document

Fingerprint

Projects

ALGO: Autonomous Live Game Operations

Cite this