A Framework for Composition and Reuse on the Linked Open Data

In recent years, many linked open datasets have been published, enabling data access and interoperability at a new scale. However, reusing rules, queries and processes is still difficult: applications are usually developed from the ground up, reinventing queries, inferences and operations that others might have created before. To address this issue, we introduce reusable inference modules, created following Semantic Web standards, which make it easier to reuse inferences and calculations based on these data. These modules act simultaneously as consumers and publishers, consuming data from one or more sources and publishing results as new, derived datasets. Their internal logic is encapsulated to simplify application development and developers need only configure rules and queries.