2015
DOI: 10.1109/tciaig.2015.2401979
|View full text |Cite
|
Sign up to set email alerts
|

Churn Prediction in Online Games Using Players’ Login Records: A Frequency Analysis Approach

Abstract: The rise of free-to-play and other service-based business models in the online gaming market brought to game publishers problems usually associated to markets like mobile telecommunications and credit cards, especially customer churn. Predictive models have long been used to address this issue in these markets, where companies have a considerable amount of demographic, economic, and behavioral data about their customers, while online game publishers often only have behavioral data. Simple time series' feature … Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
16
0
4

Year Published

2018
2018
2023
2023

Publication Types

Select...
7
2

Relationship

0
9

Authors

Journals

citations
Cited by 55 publications
(20 citation statements)
references
References 16 publications
0
16
0
4
Order By: Relevance
“…The problem of churn prediction has been tackled in many different domains such as telecommunication [10][11][12][13][14][15], banking [16,17], subscription services [18], game businesses [19], and retailing [20]. In general, most of the efforts in churn studies involve prediction with different definitions of the churn event [14,21], evaluating new data mining algorithms [15,22,23], and introducing ways to deal with large volumes of data [24].…”
Section: Introductionmentioning
confidence: 99%
“…The problem of churn prediction has been tackled in many different domains such as telecommunication [10][11][12][13][14][15], banking [16,17], subscription services [18], game businesses [19], and retailing [20]. In general, most of the efforts in churn studies involve prediction with different definitions of the churn event [14,21], evaluating new data mining algorithms [15,22,23], and introducing ways to deal with large volumes of data [24].…”
Section: Introductionmentioning
confidence: 99%
“…Since different combinations of data have different analytical power, it is necessary to determine most suitable data for the analysis being performed. RFM features (Recency, Frequency and Monetary) proved to be a good source of customer behavior in general [3,12], but also in the F2P online gaming market [6,27]. Recency features are related to the time of the last usage of the service; frequency features are related to how often the service is used, while monetary features are related to the total money that the customer has spent on services over a certain period.…”
Section: Data Transformationmentioning
confidence: 99%
“…Free-to play (F2P) games and other service-based business models are dominant in the online gaming market today. Within the F2P business setting, instead of charging a single upfront fee for the game license, companies make money with subscriptions, advertising, or microtransactions from paying players throughout their lifetime as customers [6]. F2P games are free to download and the developer hopes to generate revenue via in-app purchases and/or advertising [20].…”
Section: Introductionmentioning
confidence: 99%
“…The results from the theory‐driven model significantly outperformed those from diffusion‐based churn prediction; the experimental results also showed that achievement and socialization motivations are key factors to churning. A frequency analysis approach was adopted for churn prediction in online games using players' log‐in records (Castro & Tsuzuki, ). Kawale, Pal, and Srivastava () proposed a modified diffusion model to propagate the influence vector in a player's network, which represents the social influence of the network on the player.…”
Section: Related Workmentioning
confidence: 99%