Abstract: Rulesets give a powerful knowledge capture formalism, but many problems resist full formalization. So we should aim for partial success. Rules are a means of knowledge capture, a next step after schemas and ontologies. But to reach their potential, we must devise ways to specify and operate “imperfect” systems, assisted by humans.

 

This talk describes three ways that rulesets can partially automate customers’ routine tasks to support incremental and agile systems: 1) use a mix of rules, encapsulated English prose, and stored answers to express complex policies; 2) principled ways to automate part of the run time processing, so that rulesets can be flexible about what information they receive without re-programming; and 3) explaining to users what rules apply in their situation

 

Compliance is a natural initial application; eventually the approach might also be used for regulations and laws.

Short Bio: Arnon Rosenthal has consulted and published in data sharing and administration, databases, clouds, data security, policy based systems, and graph algorithms. He has (according to ResearchGate) 150+ publications, and 4000+ citations. His work tries to address many sides of a problem simultaneously, clarify and decompose the challenges, understand the pragmatics, simplify, and generalize components of a solution, for a realistically imperfect world. He has worked at The MITRE Corporation, and Computer Corporation of America, and Sperry Research, was a visiting researcher at IBM Almaden Research and ETH Zurich, and a faculty member at University of Michigan (Ann Arbor). He holds a Ph.D. from the University of California, Berkeley.