Securing Cloud Data: An Approach for Cloud Computing Data Categorization Based on Machine Learning

Introduction/Importance of Study: A novel innovative technique known methodical approach is referring as cloud computing (CC), which allows users to store data on remote servers that are accessible through the internet. This method makes it simple to move and retrieve vital and personal data storage. As a result, the demand for it is rising daily. This can be used to store a variety of data, including multimedia content, paperwork-based files, and financial transactions. Furthermore, by lowering operating and maintenance expenses, CC lessens the reliance of the services on local storage. Novelty statement: Current systems apply only one key size with which all data is encrypted without concerning the level of privacy of the data. This results in higher processing costs and longer processing times. Furthermore, none of these methods improves secrecy and only achieves a low accuracy rate in data classification. Material and Method: This study presents a cloud computing strategy for data sensitivity that is based on automated data classification. The model suggested in this study utilizes Random Forest (RF), Naïve Bayes (NB), k-nearest neighbor (KNN), and support vector machine (SVM) classifiers to achieve automated feature extraction. This methodology is designed to operate effectively across three sensitivity levels: basic, confidential, and highly confidential. Results and Discussion: The experiments were performed on the Reuters-21578 dataset, which consists of 21,578 documents. The simulation results demonstrated that the three proposed models achieved accuracy rates of 97%, 96%, and 95%, respectively. These findings indicate that SVM, RF, and KNN outperform NB in classification performance. Concluding Remarks: Additionally, the suggested study offers helpful recommendations for researchers and cloud service providers (like Dropbox and Google Drive).