To be or not to be...social

Abstract

Mobile games make up the largest segment of the games industry, in terms of revenue as well as players. Hundreds of thousands of games are available with most being free to download and play. In freemium games, revenue is predominantly generated by users making in-game purchases. As only a small fraction of users make purchases, predicting these users and their Customer Lifetime Value are key challenges in Game Analytics and currently barely explored in academic research. Furthermore, while social factors have been shown to be essential for retention in games in general, the impact on retention and monetization in mobile games is unexplored. In this paper, the problem of de ning social features in freemium casual mobile games is addressed through a case study with over 200,000 players. e study evaluates the in uence of speci c types of social interactions typical of casual mobile games, on predictions of premium users and Customer Lifetime Value by applying classi ers and regression models respectively. Results indicate that social activity Permission to make digital or hard copies of part or all of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for pro t or commercial advantage and that copies bear this notice and the full citation on the rst page. Copyrights for thirdparty components of this work must be honored. For all other uses, contact the owner/author(s).

Cross-Domain Adaptative Learning for Online Advertisement Customer Lifetime Value Prediction

“…Traditional machine learning methods commonly leverage tree-based algorithms such as Extreme Gradient Boosting (XGBoost) (Chen and Guestrin 2016). Drachen et al (Drachen et al 2018) leverage both Random Forest (Breiman 2001) and XGBoost algorithms to predict mobile game customer LTV with social feature involved. Besides, Vanderveld et al (Vanderveld et al 2016) design a model based on Random Forest with engagement features collected from a e-commerce platform to predict future value of an individual customer.…”