With the widespread use of mobile phones and smartphone applications, protecting one's privacy has become a major concern. Because active defensive strategies and temporal connections between situations relevant to users are not taken into account, present privacy preservation systems for cell phones are often ineffective. This work defines secrecy maintenance issues similar to optimizing tasks, thereby verifying their accuracy and optimization capabilities through a hypothetical study. Many optimal issues arise while preserving one's privacy and these optimal issues are to be addressed as linear programming issues. By addressing linear programming issues, an effective context-aware privacy-preserving algorithm (CAPP) was created that uses an active defence strategy to determine how to release a user's current context to enhance the quality of service (QoS) regarding context-aware applications while maintaining secrecy. CAPP outperforms other standard methodologies in lengthy simulations of actual data. Additionally, the minimax learning algorithm (MLA) optimizes the policy users and improves the satisfaction threshold of the contextaware applications. Moreover, a cloud-based approach is introduced in the work to protect the user's privacy from third parties. The obtained performance measures are compared with existing approaches in terms of privacy policy breaches, context sensitivity, satisfaction threshold, adversary power, and convergence speed for online and offline attacks.