Online game player churn prediction based on multiplex social influence through network embedding
Ming Guo, Yan Liu · 2026
Read Online game player churn prediction based on multiplex social influence through network embedding by Ming Guo, Yan Liu (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/plos-pone-0356200.txt.
Description
Ming Guo and Yan Liu predict which online game players are about to quit, using the social fabric of the game itself, in this 2026 study. Most churn models look only at a player's own behavior, yet players are tied to each other in overlapping ways. The authors built two networks from a real game: declared friendships and the implicit ties between teammates who fight computer opponents together.
They compressed each network into numerical fingerprints that capture a player's position at several scales, then combined those with behavior data for classification. The combined model predicted churn with accuracy above ninety percent and clearly beat models using only one kind of relationship. Extra tests confirmed the two social layers reinforce each other. For game operators, the message is that a player's friendships and teammates signal their loyalty as loudly as their play habits.
Catalog record
| CREATOR | Ming Guo, Yan Liu |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 87889 |
| LICENSE | CC_BY |
| RIGHTS | CC BY |
| SOURCE | PLOS |