The PNNL Streaming HYpothesis REasoning (SHYRE) Use Case

We will introduce three national lab use cases that require efficient reasoning on large and diverse data streams. Attendees will see examples of challenges including sampling rates and cache maintenance, deductive reasoning with partial information. We will introduce a semantic web framework that addresses challenges around (a) choosing assertions worthy of preserving within a fixed-size cache and (b) performing deductive reasoning continuously using a "smart cache" in order to effectively draw relevant conclusions as quickly as possible. The use of a smart cache differentiates our work from state-of-the-art works in deductive stream reasoning in that the cache enables us to temporarily store propositions that are no longer in the stream window.

We will present a detailed first hand case study of our prototypes that use 3 different semantic web reasoners and describe the benefits in concrete application terms, thereby providing testimonials for the approach.

A framework that creates an in-memory cache and uses the Pellet reasoner to perform OWL-DL, SPARQL queries and works in the domain of mass spectroscopy
A framework that creates the cache within a StarDog triple store, and demonstrates the challenges when querying over a network during stream ingestion
A framework that uses AllegroGraph as a cache, and highlights RDFS++ reasoning on the data stream along with its benefits and challenges

William Smith is a Research Engineer at the Pacific Northwest National Laboratory and active in initiatives supporting Linked Open Data, Semantic Web Technologies and Deductive Logic modeling. His research focus includes: (1) supporting the development of the Semantic Web and web ontology modeling and (2) authoritative data reuse and interlinking across research communities.

Deborah L. McGuinness is the Tetherless World Senior Constellation Chair and Professor Computer Science and Cognitive Science at Rensselaer Polytechnic Institute as an endowed chair of the Tetherless World Research Constellation. She is one of the creators of the OWL Web Ontology Language. McGuinness was previously the Director and Senior Research Scientist at the Knowledge Systems, Artificial Intelligence Laboratory at Stanford University. She is a leading expert in knowledge representation and reasoning languages and systems and has worked in ontology creation and evolution environments for over 20 years. Deborah is well known for her leadership role in semantic web research and applications of semantic web technology, particularly for environmental and health informatics applications.