Abstract:
With the growing adoption of cloud computing and the increasing popularity of digital technologies, personal data storage
and processing in cloud environments has become essential. However, as organizations and individuals embrace the benefits of cloud
services, the security of personal sensitive information within this dynamic ecosystem has become a top priority. Ensuring the
confidentiality, integrity, and availability of personal data in the cloud is critical to mitigate the risks associated with cyber threats. As
cyber threats continue to evolve, it is essential to adopt innovative approaches to ensure personal data security measures. This article
introduces a new approach, leveraging machine learning, and data masking techniques, using both serverless and secret vault services
provided by most of cloud service providers (CSPs). Data masking techniques are employed to further protect sensitive information
from unauthorized access. This paper explores and assesses the effectiveness of machine learning algorithms including LSTM, CNN,
and MLP in classification tasks. The results highlight CNN's outstanding performance, achieving a remarkable 100% accuracy. This
ensures perfect classification with double validation using the pattern matching technique. Furthermore, an analysis of download and
upload time costs reveals that data processing using our model has no significant impact on the execution time metric.
