Enhancing cloud network security with a trust-based service mechanism using k-anonymity and statistical machine learning approach

Abstract

This Research work addresses the pressing need within cloud computing for a trust-based service mechanism that effectively manages the burgeoning volume and variety of data while mitigating privacy concerns. The primary aim is to address pressing security challenges within cloud networks through a novel approach tailored to enhance privacy preservation mechanisms. Experiments were done on a variety of datasets using a hybrid privacy-preserving strategy to assess the efficacy of the suggested solution. The datasets were divided into both testing and training sets for the experimental design, using a 70% validation ratio for training. The method's performance was compared with that of existing strategies, including caching and spatial K-anonymity (CSKA) and privacy-preserving incentive and rewarding (PPIR), using precision, recall, and F-measure analysis. The findings show that the suggested strategy performs better than the baseline approaches in a number of assessment measures, indicating its greater capacity to protect privacy in cloud environments. Specifically, the approach achieved an average precision of 0.85, significantly surpassing the precision values of existing techniques by 8-10%. Moreover, the method exhibited an average recall of 0.84, indicating its robustness in recalling values across all test samples. Across various experiments, our method consistently achieved impressive F1 scores ranging from 0.80 to 0.85, underscoring its robustness in maintaining a balance between precision and recall. Furthermore, with an accuracy hovering around 0.85, our approach demonstrated remarkable proficiency in accurately classifying instances while preserving privacy in cloud environments. These promising results underscore the efficacy of the proposed approach in enhancing privacy preservation mechanisms within cloud networks, paving the way for more secure and reliable cloud computing infrastructures. By leveraging a hybrid privacy-preserving method, the paper offers a holistic approach to address the complex problems faced by cloud networks in safeguarding sensitive information. The experimental evaluation demonstrates the efficacy of the proposed approach, highlighting its superior performance compared to existing techniques.