Securing Cloud Data: A Machine Learning based Data Categorization Approach for Cloud Computing

Cloud computing (CC) a new systematic model that allows users to store data on remote servers that are accessible via the internet. Due to this approach, the personal and essential data are being stored with easy to access and move. Because of this, it demands is increasing day by day. On this, one can store different data such as financial transactions, paperwork based files and multimedia content. Not only this, but CC also reduce the services dependency on local storage by reducing operational and maintenance costs. Existing systems such as encrypt all data with the same key size, regardless of the data's level of confidentiality due to which the processing cost and time become increased. Moreover, all such techniques classify the data with a low accuracy rate and don’t provide better confidentiality. In this research, a cloud computing approach based on automated data classification has been presented for data sensitivity. The proposed model is based on three level of sensitivity i.e. basic, confidential and highly confidential using Random Forest (RF), Naïve Bayes (NB), k-nearest neighbor (KNN) and support vector machine (SVM) classifiers for training the proposed model with automated feature extraction. The proposed model achieved 92% accuracy that has been showed in simulation results. From this, we conclude that RF, NB, KNN performs better than SVM. The proposed research also provides useful guidelines for cloud service providers (e.g., drop box and Google drive) and researchers.