Smarter interoperability based on automatic schema matching and intelligence amplification

The process of connecting enterprise systems and applications to digital platforms and ecosystems requires interoperability on different levels and eventually leads to the task of creating a schema mapping. At present, this task is carried out manually and prior research has proven this task difficult to automate. In this paper, we discuss the suitability of machine learning approaches to create an auto-mapping functionality, so different schemas and standards can (partially) be mapped automatically, and incorporate a symbiotic approach to improve the matching result. To the best of our knowledge, this is a new approach with potential to reduce the time needed for schema matching tasks. The main contribution of this paper is a reference architecture and prototype for smarter interoperability using a combination of automatic schema matching, based on machine learning, and intelligence amplification