SemMat: Federated Semantic Services Platform for Open materials Science and Engineering

Abstract : An open source framework was developed to enable crowd-sourcing and curation of controlled vocabularies. The framework was then applied to the materials and manufacturing domain to create several hundred terms and definitions extracted from various structured sources. The domain model and the curation platform supports preserving important metadata information including provenance. Additionally, a novel Singleton property approach was implemented to enable representation of relevant information efficiently and in a semantically clean manner. A visualization tool, iExplore, was developed to provide the capability to visually search and explore links between materials and manufacturing domain concepts. Tools and techniques were developed to enable identification of materials entities in unstructured data sources and documents (such as PDF documents).