Personalized web search (PWS) has been shown to improve the quality of online search services. According to the findings, a fundamental hurdle to the widespread adoption of PWS is users' reticence to share private information while conducting searches. Our research focuses on how PWS (Personalized Web Services) applications that use hierarchical user profiles to express user preferences protect user privacy. We introduce UPS, a brand-new personalized web search (PWS) platform, in this study. Using adaptive query-based approaches, this system is aimed to efficiently generalize user profiles while adhering to user-defined privacy constraints. Our runtime generalization seeks to strike a balance between two predictive metrics that assess the value of customisation as well as any privacy implications connected with profile sharing. We provide the greedy DP and greedy IL algorithms for runtime generalization. In addition, we offer an online tool for estimating the possible benefits of query personalisation. A variety of in-depth tests have proven the effectiveness of our approach. Furthermore, the investigation's findings show that GreedyIL is substantially more efficient than Greedy DP..