Mobile game advertisement strategy dynamic optimization method based on double fatigue degree model
By constructing a virtual player group simulation environment with a dual fatigue model, and combining AI agents and systematic search algorithms, the problem of inaccurate churn prediction due to advertising interference in mobile games was solved. This enabled automated optimization of level and advertising strategies, improving game operation efficiency and commercial revenue.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-31
- Publication Date
- 2026-07-03
AI Technical Summary
Existing technologies in mobile games fail to effectively consider the impact of advertising interference on player churn, resulting in inaccurate churn predictions, an inability to dynamically optimize the balance between advertising strategies and level difficulty, and a lack of automated optimization methods.
We construct a virtual player group simulation environment based on a dual fatigue model, use AI agents to evaluate level difficulty and quantify ad fatigue, and combine a dynamic eCPM model and a systematic search algorithm to automatically optimize level and ad strategies to maximize the predicted total lifetime value.
It improves the accuracy of player churn prediction, enables automated collaborative optimization of level difficulty and advertising strategies, balances player experience and commercial revenue, reduces reliance on manual testing, and improves game operation efficiency.
Smart Images

Figure CN121810349B_ABST
Abstract
Citation Information
Patent Citations
Game level generation recommendation method, system, equipment and medium
CN120617962A
Dynamic game difficulty self-adaptive adjustment method and system based on user behavior feedback
CN120789656A