MongoDB-Based Modular Ontology Building for Big Data Integration

Big Data are collections of data sets so large and complex to process using classical database management tools. Their main characteristics are volume, variety and velocity. Although these characteristics accentuate heterogeneity problems, users are always looking for a unified view of the data. Consequently, Big Data integration is a new research area that faces new challenges due to the aforementioned characteristics. Ontologies are widely used in data integration since they represent knowledge as a formal description of a domain of interest. With the advent of Big Data, their implementation faces new challenges due to the volume, variety and velocity dimensions of these data. This paper illustrates an approach to build a modular ontology for Big Data integration that considers the characteristics of big volume, high-speed generation and wide variety of the data. Our approach exploits a NOSQL database, namely MongoDB, and takes advantages of modular ontologies. It follows threemain steps: wrapping data sources toMongoDB databases, generating local ontologies and finally composing the local ontologies to get a global one. We equally focus on the implementation of the two last steps.