Data mining is used in many contexts, such as business, healthcare, science, and government. Data mining frequently makes use of categorization. Recent years have seen a meteoric growth in the development of both theoretical and practical classification systems in response to rising privacy concerns. These changes were implemented using a wide variety of security frameworks. As cloud computing grows in popularity, more and more people are turning to it for secure data storage and transmission to remote servers for data mining. Data encryption in the cloud renders obsolete the existing privacy protection classification methods. The purpose of this study is to identify a solution to the issue of encrypted data classification. For encrypted cloud-stored data, we recommend utilizing a powerful k-NN classifier. The suggested protocol is designed to protect the confidentiality of users' search histories and access logs. To our knowledge, this is the first research to employ the semi-honest model in order to develop a trustworthy k-NN classifier capable of processing encrypted data. Furthermore, we use an observational dataset with multiple adjustment factors to evaluate our proposed approach in the actual world.