Due to the growing demand for data integration in different fields, it has led to the development of innovative methods for the semantic integration of heterogeneous data sources. This study proposes a new approach that uses the Resource Description Framework (RDF) and Relational Database (RDB) in the RDB to RDF Mapping Language (R2RML) for the semantic integration of data resources within the framework of an integrated semantic model. This approach consists of an Extract, Transform, and Load (ETL) pipeline connected to an RDF triple store that allows the use of multiple ontologies in different domains and the management of distinct data models. The applicability of this method was demonstrated in the fields of healthcare and the Internet of Things, enhancing a unified view of data and interoperability. This process involves converting the data to RDF, creating an integrated RDF specification, storing it in an RDF repository, and querying it through a SPARQL endpoint, enabling intelligent decision-making processes. The implementation and results of this method show the integrity of semantic data and its strength in addressing the complex requirements of semantic interaction in multi-domain environments. This research promotes the applicability of semantic technologies for integrated data management by integrating a comprehensive set of tools and proving its practical applications. The proposed innovative approach provides a promising solution for the semantic integration of heterogeneous data sources, improving interoperability and enabling intelligent decision-making processes.