1 / 22

Visual Knowledge, Inc.

Visual Knowledge, Inc. Visual Knowledge, Inc. Fifth Semantic Inter-Operability Conference Conor Shankey CEO Oct. 10 2006 McClean, Virginia. Large scale multi-agent software systems. Agents that are rapidly modeled and evolved by large number of people. Visual Knowledge.

tsylvia
Download Presentation

Visual Knowledge, Inc.

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Visual Knowledge, Inc. Visual Knowledge, Inc. Fifth Semantic Inter-Operability Conference Conor Shankey CEO Oct. 10 2006 McClean, Virginia Company Confidential

  2. Large scale multi-agent software systems Agents that are rapidly modeled and evolved by large number of people Visual Knowledge What is our technology? Company Confidential

  3. Systems of agents that can be federated and can create executable systems Agent systems that cope with conflicting ideas, causality and context Visual Knowledge What is our technology? Company Confidential

  4. Semantic Agent • Like living Lego Pieces • Atom of knowledge, content, and behavior. • At the most granular level, everything in Visual Knowledge is made up of semantic agents. • Semantic agents are declarative specifications for services. They are not algorithms. Their DNA is knowledge — knowledge about resources, content, media, language, processes, functions, and how to communicate with other agents. • Semantic agents collaborate with other agents across platform(s) to provide services and capabilities. • Semantic agents can be modeled, built, purchased, shared, acquired, and linked together Company Confidential

  5. Inference Agents • Semantic Agents that execute inferencing primitives • Can be assembled and re-used in more complex models • Can be customized for different inferencing paradigms • OWL Description Logic • Lexical Analysis • Higher order logic • Numerical / Statistical modeling • Apriori Reasoning • Bayesian Reasoning • Spatial inferencing Company Confidential

  6. Capability Packages • A set, pattern, or assemblage of semantic agents that work together to perform a set of functions that deliver a result that meets a need. • Examples include: semantic (concept-based) search, semantic navigation, concept and relationship extraction, auto-categorization, etc. • Capabilities are building blocks of solutions. • Capabilities can be developed, purchased, shared, Company Confidential

  7. Visual Knowledge Pluggable Packages of Semantic Agent Capability via Web Services Composite Apps OWL Engine MOF/ UML/ERD ISO11179 VK Upper Ontologies Meta Meta Semantic Microkernel Company Confidential

  8. Visual Knowledge Semantic Wiki provides a scalable “user friendly layer” over models Semantic Wiki OWL Engine MOF/ UML/ERD ISO11179 VK Upper Ontologies Meta Meta Semantic Microkernel Company Confidential

  9. Visual Knowledge Domain Capabilities are pluggable and specific Domain Capability Semantic Wiki OWL Engine MOF/ UML/ERD ISO11179 VK Upper Ontologies Meta Meta Company Confidential Semantic Microkernel

  10. Domain Capability Domain Capability Visual Knowledge Domain Capabilities are pluggable and specific Semantic Wiki OWL Engine MOF/ UML/ERD ISO11179 VK Upper Ontologies Meta Meta Company Confidential Semantic Microkernel

  11. Specific Use Case - Compliance • Regulatory Compliance is becoming a tremendous burden for large companies ( ~ 20% of costs in large banks) • Most policy is represented in text in documents • Policies are opaque, ambiguous and require humans to manually read to interpret - subjective • Gaps/Conflict between one document and the next • Policy needs to be re-interpreted or applied to each level of implementation • Change in policy creates extraordinary inertia Company Confidential

  12. Specific Use Case - Compliance • Regulatory Compliance is becoming a tremendous burden for large companies ( ~ 20% of costs in large banks) • Most policy is represented in text in documents • Policies are opaque, ambiguous and require humans to manually read to interpret - subjective • Gaps/Conflict between one document and the next • Policy needs to be re-interpreted or applied to each level of implementation • Change in policy creates extraordinary inertia Company Confidential

  13. Specific Use Case - Compliance • Regulatory Compliance is becoming a tremendous burden for large companies ( ~ 20% of costs in large banks) • Most policy is represented in text in documents • Policies are opaque, ambiguous and require humans to manually read to interpret - subjective • Gaps/Conflict between one document and the next • This means that the interpretation is wrong • Policy needs to be re-interpreted or applied to each level of implementation • Change in policy creates extraordinary inertia Company Confidential

  14. Compliance Pain Points • Paper based • Continuous Regulatory change • Existing system silos don’t understand cross organization compliance issues • Human error • Forgetfulness • Inconsistent interpretation • Manual communication • Code / algorithms don’t understand compliance Company Confidential

  15. Corporate Policy Auditors Business Unit Policy Testers Vertical Opacity of Policy and Reporting Standard Audit Programs Periodic Processing Testing Company Confidential

  16. Compliance Documents, Process Control Manuals RCSA Test Evidence Catalog RCSA Test Evidence Catalog Semantic Wiki Semantic Framework Ontology Direct Linkage Interfaces to LOB Applications, Content, Data Reasoning Engine Semantic Agents Compliance Designer Dashboard Policy Wiki Collaborative Environment Configuration / Administration Top-Down Visibility Policy Documents Model-Driven Agility Continuous, Consistent, Self-Assessment Policy Text into Model Compliance Concepts + Relationships Composite Application with Links to Evidence Demonstrate Linkage to Policy Roles and Responsibilities, Segregation of Duties Semantic Inference: Early Warning System for Gaps Testing Intelligent Work Management Automatic Email Notifications, Reminders, and Escalations Company Confidential

  17. VK Semantic Wiki for Compliance • Semantic Wikis are used to absorb knowledge from documents • Subject Matter experts highlight text and connect to ontology “interpret” • Ontologists evolve model of organizations, systems, tasks, etc. • Application builder evolve ontology driven application • End-user’s experience custom workflow with direct links to original policy documentation Company Confidential

  18. VK Semantic Wiki • RDF/OWL ontology driven wiki • User friendly WYSIWYG editor • Community centric knowledge visibility • All wiki concepts are represented as OWL individuals • Can be connected or inferred to OWL classes • Plug-in ontology based capabilities • FOAF based contact/admin management • Dublin core based file attributes • DOAP project/tasks • Conference/Event ontology • Other domain ontologies (compliance, PM, etc.) Company Confidential

  19. Semantic Agents Semantic Agents are the • hypertext in the wiki • elements of the ontology (model) • organization, people, tasks, systems and notifications to people • all the inferences or logic and intelligent actions • are self aware proxies of their real world counterparts Company Confidential

  20. Semantic Agents Semantic Agents • Tasks, Organization, Systems and Responsibilities expressed as semantic agents • Assignments, authority inferred by Agents via ontology and higher order inferencing agents • New responsibilities and determination of violations via ontology and higher order inferencing agents Company Confidential

  21. Semantic Agents W3C OWL/RDF standards are used extensively • All models of policy, workflow, organization, notifications • FOAF for contacts • Wiki hypertext is pure HTML with concept links Company Confidential

  22. Semantic Compliance Solution Architecture – Intelligent Business Feedback System Semantic Agent Ecosystem Company Confidential

More Related