Abstract:

Abstract Insider threats are hostile actions that a legitimate employee of a company could commit. For both commercial and governmental enterprises, insider threats pose a significant cybersecurity risk since they have a considerably greater potential to harm an organization's assets than external attacks. The majority of currently utilised insider threat methodologies concentrated on identifying common insider attack scenarios. This research propose novel technique in data leakage detection in cloud computing based on data classification using deep learning architectures. Here the input data has been collected as network data and processed for noise removal, smoothening. The classification has been done based on Generative Regression kernel SVM. The experimental findings have been calculated in terms of RMSE, SNR, F-1 score, recall, accuracy, and precision. The proposed model offers practical approaches to deal with potential bias and class imbalance issues in order to design a system that effectively detects insider data leaking. Proposed technique attained accuracy of 97%, precision of 92%, recall of 67%, F-1 score of 66%, RMSE 62% and SNR of 61%.

Keywords: