SILK: Rules, Semantic Web, and Databases Take a Leap in Scalable Logical Power
A WEBCAST FROM SEATTLE/BELLEVUE
WEBCAST
Benjamin Grosof Vulcan Inc., Seattle, WA, USA
Tags: SILK, Courteous LP, Hyper Logic Programs, Hyper LP, Argumentation Theories, Hypermonotonic mapping, HiLog LP,
|
The area of semantic rules is perhaps the most important frontier today for the semantic web's core technology and standards. Rules extend databases and ontologies with more powerful, flexible, and active forms of "structured" knowledge (as opposed to "unstructured" knowledge such as text), and have a number of close relationships to other aspects of the overall semantic web such as query/search, trust, and services. Two major initial industry standards have been finalized in the last year, including W3C Rule Interchange Format (RIF) and OWL 2 RL Profile (OWL-RL, the Rules subset of OWL). Industry investments and acquisitions in the technology are accelerating, including by IBM and Oracle -- which supports semantic web rules functionality in its core database product. The expressive abstraction level of knowledge representation (KR) in today's widely deployed kinds of structured knowledge management -- databases, business rules, and previous semantic web -- has not advanced significantly since the early 1980's. A single semantic KR -- normal/Horn LP -- underpins such previous knowledge management. (That KR was originally developed in order to formalize the semantic aspect of SQL databases and Prolog.)
AbstractWe overview and demonstrate the SILK language and system for advanced semantic web rules. SILK has a new knowledge representation (KR): *hyper* logic programs, which radically extends normal declarative logic programs (LP). Rules can be newly defeasible, higher-order / meta-, and omni-directional, as well as reactive -- yet retain computational scalability. SILK thus newly addresses critical requirements in representing: negation and exceptions; updating, revision, mapping, and merging of knowledge; process causality and state change; debate and trust; and social pragmatics of knowledge and communication. The SILK system includes a powerful efficient reasoner that answers queries, maintains views, and launches external actions (e.g., web services), as well as a GUI for querying and entering rich knowledge. SILK supports knowledge integration via web interchange with: SQL/ODBC and SPARQL/RDF databases; OWL-RL and RDFS ontologies; and RIF rules. SILK's pilot application reasons about causal hierarchical processes in college-level biology. SILK has a large variety of other potential applications, e.g., horizontally in ontological mediation, policies/regulations, and agile workflow -- and vertically in biomedical, e-commerce, privacy, defense, and mobile. SILK newly synergizes, then extends, techniques drawn from two decades of disparate previous LP research advances. Part of Vulcan Inc.'s Project Halo, SILK is a highly ambitious R&D effort that is redefining the KR playing field for business rules and rule-based process management, as well as semantic web. It addresses fundamental requirements for scaling up to widely-authored VLKBs (Very Large Knowledge Bases) in business, government, and science that answer questions, proactively supply information, and reason powerfully. (Project Halo also includes Semantic MediaWiki+, the leading semantic wiki software, which extends what Wikipedia runs on.) The SILK UI is currently being extended to facilitate usage by subject matter experts (not just knowledge engineers), in part via controlled natural language. We discuss prospects for the SILK approach to effectively interchange and integrate a high percentage of the world's structured knowledge starting from today's legacy forms. "SILK" stands for "Semantic Inferencing on Large Knowledge", what the next generation Web will be spun from.
Speaker BioBenjamin Grosof is a senior research program manager at Vulcan Inc., the parent company of Paul G. Allen (co-founder of Microsoft). There he conceived and leads a large research program in the area of rule-based semantic technologies and artificial intelligence (AI), aiming to be a game changer for knowledge representation and question answering. This is centered on developing the SILK language and system for semantic web rules that are defeasible and higher-order/meta-, yet scalable. In addition, he has a part-time expert consulting business, advising companies large and small on technology and related strategy, and is affiliate faculty in AI at U. Washington. Previously he was an IT professor at MIT Sloan (2000-2007) and a senior software scientist at IBM Research (1988-2000). He has pioneered semantic technology and standards for rules, their combination with ontologies, their application in e-commerce and business policies, and business roadmapping of the Semantic Web. He co-founded the influential RuleML industry standards design effort and prototyped it in SweetRules, the main bases for the W3C Rule Interchange Format (RIF) standard. He was lead inventor of the rule-based technique which rapidly became the currently dominant approach to commercial implementation of W3C OWL (Web Ontology Language) and the main basis of its RL (Rules Profile) standard, and of several other fundamental technical advances in knowledge representation . His background includes three major industry software releases, two years in software startups, a Stanford PhD in AI, a Harvard BA, and over 50 refereed publications. External ResourcesBenjamin Grosof, Mike Dean, Michael Kifer: The SILK System: Scalable and Expressive Semantic Rules, Poster Session ISWC 2009 SILK demo video (biology - initial version of 7 September 2010) Benjamin Grosof: Semantic Rules on the Web, a tutorial held at ISWC 2009 |
Webcast here Date 9/29/2010 Time 6pm PST/9pm EST |