A Semantic Metadata Framework for Enhanced Accounting Information Systems: Design, Implementation, and Performance Analysis

This paper presents a new way of using semantic metadata for accounting information systems that enriches financial transactions with context of business information. The framework has a four-layer architecture consisting of base transaction layer, business context layer, relationship layer, and compliance layer. This framework helps to capture and use the context of business in accounting. Our framework is practically validated based on JSON-LD implementation and extensive simulation tests. The simulation in our paper, which consisted of one million transactions, yielded a query response time of 5.9 milliseconds with 100% consistency. The system's continuous processing rate was 44 transactions per second (TPS) with cache usage efficiency being 71.05%. The findings show that practical performance characteristics can be maintained in accounting systems despite semantic metadata integration. The framework allows for enhanced business intelligence, automated compliance checking, and improved decision support in accounting systems. Our tests demonstrate that semantic details can be seamlessly incorporated into classic accounting practices without hindering the performance of the system or losing data.