Context-Aware Access Monitoring With Tracking and Alerting for Cloud-Based Invoice Processing In Financial Automation

The proactive strategy of detecting vulnerabilities and ensuring secure processing of invoices in the cloud is termed Risk Management (RM). Yet, the existing studies didn’t automatically adjust the access privileges according to contextual risk along with tracking and alerting, affecting the model. Thus, this paper presents context-aware access monitoring with tracking and alerting for cloud-based invoice processing in financial automation using Parabolic Ramp Fuzzy (PR-Fuzzy). Primarily, for accessing the invoice, the admin creates the role and access policy. Next, by using Polyphase Lifting Discrete Wavelet Transform (PLDWT), the invoice data is watermarked. Afterward, a digital signature is created for the watermarked invoice data based on the Inverse Deterministic Salt-based Digital Signature Algorithm (IDS-DSA). The digital signature is verified at the destination side, and watermarking is removed for editing the invoice. The attributes are extracted using Optical Character Recognition (OCR) from the edited invoice. Next, by using SwAT Graph Toeplitz Convolutional Networks (SGTCNs), invoice fraud detection is performed. The access is blocked if fraud is detected; otherwise, the edited invoice is watermarked, authorised, and sent to another department. Lastly, it reaches the user. Similarly, by using PR-Fuzzy, the context-aware access monitoring is performed. The access is blocked if high risk is obtained, and an alert is sent to the admin. Similarly, by using Exponential-GINI-BLOOM based Merkle-Tree (EGB-MT), the tree is constructed with invoice accessing information and then shared with the admin for tracking and monitoring. As per the results, the proposed model took a lesser rule generation time of 1024ms.