Intelligent medical data analytics is only one of several cloudbased IoT machine learning applications that rely on phrase search to retrieve documents that include an exact term. Data owners often encrypt documents (such as medical records) before outsourcing them to the cloud to prevent the leakage of sensitive information. On the other hand, this makes the search operation a very difficult process. Smart encrypted data processing in the cloud is the focus of this article, where we provide P3, a privacy-preserving phrase search strategy that is both efficient and effective. We find the geographical association of numerous requested keywords over encrypted data by using the homomorphic encryption and bilinear map in our technique. In addition, it safeguards users' search habits by use of a probabilistic trapdoor creation method. A comprehensive evaluation of P3's security features reveals the benefits of these enhancements. We put a prototype into action and do compreh