INTELLIGENT PAYROLL-TO-GENERAL LEDGER RECONCILIATION USING MACHINE LEARNING-BASED DISCREPANCY DETECTION

Abstract

Payroll-to-General Ledger (GL) reconciliation is a critical financial control process that ensures accuracy, transparency, and compliance between employee compensation records and enterprise accounting systems. Traditional reconciliation approaches rely heavily on manual verification, predefined accounting rules, and periodic audits, which are inefficient for large-scale organizations with complex payroll structures. This paper proposes an Intelligent Payroll-to-General Ledger Reconciliation framework using Machine Learning-Based Discrepancy Detection by integrating machine learning, automated data pipelines, enterprise resource planning (ERP) systems, anomaly detection, and intelligent financial analytics. The proposed methodology analyzes payroll transactions, accounting entries, historical reconciliation patterns, and financial attributes to identify inconsistencies, classify discrepancies, and recommend corrective actions. Experimental evaluation demonstrates improvements in reconciliation accuracy, anomaly detection capability, processing efficiency, audit readiness, and financial control effectiveness. The proposed framework provides a scalable and intelligent solution for automating payroll reconciliation processes while improving enterprise accounting reliability and reducing operational risks. Keywords— Payroll Reconciliation, General Ledger, Machine Learning, Discrepancy Detection, Financial Analytics, ERP Systems, Anomaly Detection, Intelligent Automation.