A Game Theoretic Approach toward Privacy Preserving for Mobile Learning Data Sharing

Diversified software and interactive modes have enhanced users' learning experience in mobile learning(M-Learning). Mobile learning have gradually become a new way for users to participate in learning. Users disclose personal learning data to the mobile learning platform to obtain better service quality, and on the other hand, they worry about their privacy being leaked and violated. The platform can analyze user data and provide personalized services based on it. However, the platform may sell the user's information to some malicious adversaries in order to make profits, or be attacked by adversaries that cause the privacy of users to be leaked. In actual application scenarios, the behaviors of users, platforms, and opponents in a mobile learning system affect each other, so their interests are also mutually restricted. This paper proposes a game theoretic approach to analyze the mobile learning system, and constructs a mixed strategy game model with three players: user, platform and adversary. Through the analysis of participants, behaviors, strategies, states, and payoffs, it defines the player's profit formula under mixed strategies. Finally, based on the Nash Equilibrium existence theorem, the rationality of the existence of the player's best profit is analyzed.