ASIF: A Semantic Integration Framework for Heterogeneous Big Data Environments

The increasing diversity of data generated across enterprise, cloud, and distributed computing environments has created significant challenges for integrating heterogeneous data sources into unified analytical platforms. Differences in data models, schema structures, and semantic representations often limit the effectiveness of conventional extract-transform-load (ETL) approaches. This paper presents ASIF (A Semantic Integration Framework), a scalable architecture designed to support semantic integration across structured, semi-structured, and unstructured data sources. The framework combines ontology-driven schema reconciliation, entity resolution, and distributed consistency mechanisms to enable reliable integration in large-scale data environments. ASIF was evaluated using datasets from government open-data repositories, healthcare interoperability benchmarks, and e-commerce transaction systems comprising approximately 4.2 billion records. Experimental results demonstrate that the framework significantly reduces schema conflicts while achieving high entity-matching accuracy across heterogeneous data sources. The proposed architecture also supports scalable ingestion and query processing in distributed environments. These findings indicate that semantic integration approaches can effectively address structural and semantic heterogeneity while maintaining the performance requirements of modern big data platforms. The framework provides a practical foundation for enterprise and cloud-based data integration systems operating at scale.