1. Semantic Wikis for Information Management
Working Group
Conor Shankey
CEO
March 9, 2006
Visual Knowledge, Inc.
11/27/2018

2. CEO Conor Shankey Founder & CEO of Visual Knowledge, 15 years.
Architected largest corporate wan/lan of it’s kind outside US in 1990
Architected enterprise transactional frame system to supplement/replace mission critical mainframe systems of large power utility
CTO/co-founder of several spin out companies from Visual Knowledge
Co-founder Clera Pharmaceuticals, small anti-psychotic drug discovery company
NCOR technical committee, Chair of 2006 OWL workshop (Intl Semantic Web Conference)
11/27/2018

3. Visual Knowledge What is our technology?
Large scale multi-agent software systems
Agents that are rapidly modeled and evolved by different groups of people
11/27/2018

4. Visual Knowledge
R&D through real-world implementations
11/27/2018

5. Visual Knowledge What is our technology?
Systems of agents that can be federated and can create executable systems
Agent systems that cope with conflicting ideas, causality and context
11/27/2018

6. 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
11/27/2018

7. 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
11/27/2018

8. 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
11/27/2018

9. Semantic Micro Kernel Very light weight / portable web service
Designed to support 100M+ concepts, billions of triples
Ontologies and Knowledge Base reside in Kernel
All agents, attributes, relationships reside in pure form between disk and memory system
Almost all inferencing primitives reside in kernel
API generator creates common interface for Java, C#, Python, Ruby, Smalltalk and Lisp
Ported to Windows, Linux, Mac, Super Computers, etc.
11/27/2018

10. Other Architectural Keys
All agents are first class active objects
Intelligent Semantic Change Management
Pluggable, distributed semantic capability servers
100% compatibility with OWL/RDF
11/27/2018

11. The original wiki idea
“A web site where anybody could create/edit a web page”
Structure
is not pre-determined
invented & evolved by community
neither top down or bottom up
Quick collaborative writing
Non-linear Hypertext
11/27/2018

12. Simplicity Additional Notions Very simple markup for authors
Any page can be immediately revised assuming you have the right privileges
All changes are audited and transparent to the community
“Concepts” in text can immediately become active resources (pages/links)
Simplicity
articles
11/27/2018

13. Benefits of the wiki idea
Distinct concepts or topics are built on the fly
Discourse forms around or in the context of a topic
Eliminates serialized document work flow
Team or community members can immediately see commentary in the context of a topic
Consensus
Agility
Cohesion
Speed
11/27/2018

14. version? Compared To… Each person edits a copy of the document
A poor soul merges the results
Expensive file shares
ing bulky documents
“Versions” of opaque documents everywhere
“Organizing” documents in hierarchal file system
version?
11/27/2018

15. Greatest Strength and Weakness
Topics or concepts lack semantics
A WikiWord is just a WikiWord
A page with related formatted text and WikiWords
Authored, versioned content
Instance based security
Arbitrary structure
Quick and open architecture and adoption led to lack of standardization
Security?
11/27/2018

16. Semantics to the rescue
Phrases having different meanings in different contexts
Water on mars?
Food
Space Exploration
11/27/2018

17. Semantics to the rescue
Different concepts have different properties
Spatial properties
Unique Relationships
Security is critical concern!
Some information should be presented in views to different audiences (like your social security number)
For mission critical systems, staging is essential, we need separate, federated
Development servers
Data Staging servers
QA servers
Simulation servers
Production servers
11/27/2018

18. VK Semantic Wiki Wiki concepts are just elements of an ontology
Now wiki concepts have formal properties
Enables semantic searching
Security becomes deeply integrated into wiki
Customized Wiki Views
Driven by ontology
Driven by library of visual templates
Integrated change management
Federated semantic wikis
11/27/2018

19. VK Semantic Wiki
Semantic Wikis are organized around Communities of Interest
Versions of ontologies of interest drive capabilities in Wiki
Protected worlds with controlled access to outside communities
All contacts, concepts, layouts presented in W3C standards
OWL
RDF (FOAF, Dublin Core)
HTML
11/27/2018

20. Demo Semantic Wiki OWL Driven App cshankey@visualknowledge.com
If you would like to beta, please contact
11/27/2018