MUMBAI, India, Sept. 14 -- Intellectual Property India has published a patent application (202641105041 A) filed by Anil Chandra Rajendra Pallikonda on September 01, 2026, for Artificial Intelligence-Enabled System For Detecting Anomalous Transactions In Enterprise Accounts Payable And Accounts Receivable Modules.
Inventors include Anil Chandra Rajendra Pallikonda; Naveen Komulwad; and Rajesh Kumar Reddy Kallapari.
The application for the patent was published on September 11, 2026, under issue no. 37/2026.
Abstract: The invention provides an artificial intelligence-enabled system for detecting anomalous transactions across enterprise accounts payable and accounts receivable modules. Source connectors receive timestamped transaction, master-data, approval and payment events. A canonical processor resolves transaction and entity keys, suppresses duplicates and qualifies event time, schema and freshness. An incremental store updates affected temporal states and transaction-entity graph neighbourhoods. A hybrid inference engine generates temporal and graph anomaly evidence. A validity evaluator calibrates uncertainty and removes predictive authority for unsupported inputs or operating regimes. A policy engine combines valid evidence with deterministic authorisation, segregation-of-duty and transaction- state checks. A constrained orchestration interface dispatches a reversible or approval-gated instruction with an idempotent token. A feedback engine verifies connector acknowledgement and subsequent enterprise state before updating anomaly status. The architecture reduces redundant ledger processing, repeated alerts and stale actions while preserving auditable fallback and model-governance controls.
Disclaimer: Curated by HT Syndication.
