[Editor's Note: This week, we welcome Yefim "Jeff" Zhuk of Sallie Mae as he presents a series on Knowledge-Driven Architecture. This series follows up the author’s presentation at the recent international 2011 Semantic Technology Conference San Francisco and further expands on the subject of integrated software and knowledge engineering, originally described by Mr. Zhuk in the book “Integration-ready Architecture and Design.” Part I | Part II | Part III]

Part IV – Creating a semantically rich service environment locally and across industry

Part III focused on the Conversational Semantic Decision Support (CSDS) and related Use Cases.

This example can be expanded from requirements to design and development phases, including hints on service names and application messages. Standards, recommendations and best practices offered by W3C [6] can serve as the base for conversational scripts, which would help a SME, (in this case, a software developer) to successfully implement them and create a truly semantically rich SOA environment.

In the semantically rich environment, there is no need for complex tools. The names and messages are self-explanatory and directly tied to the ontological execution model. For example, service names and operations can be consistently composed with the Actor, Action, and Object vocabulary.

Application messages can be done in JSON style:

{“time”: “currentTime”, “application”: “BestVendorApplication”, “service”: “CustomerEnterOrder”, “error”: “database is down”, “recoveryAction”: “restart database”, “notify”: “currentlyOnCallPerson”}

If this is consistent across the industry, vendors will create smaller, smarter and less expensive semantic sensitive tools to monitor and manage service operations. The same message will become a valuable record in the problem archive. Such records can be RDF-formatted to and processed to compose the “situation” and find root cause factors.

A very similar thing can be done with service messages directed to an enterprise service bus.

A semantic message listener will not need to be programmed to each specific message’s content and format. Instead, it will be able to listen to much wider themes, which can be quickly redefined by business, providing a new dimension for event-driven architecture. People-readable service messages will be quietly translated to RDF to accommodate the upcoming RDF Services standard [7].

Service monitoring and managing is one of the most dramatic areas of IT. It is also one of areas with the greatest potential to decrease cost and cut extra layers. We will cut down the need for the magic of writing monitoring scripts. Application messages can instruct applications on self-checking and self-healing.

A semantic listener will execute instructions written by application developers, who know best what to expect, how to prevent, and recover. A semantic listener will be able to combine multiple application messages into a situational scenario, something that we desperately need and completely lack today.

This is not just a technical effort. This effort requires the guidance and support from Information Governance locally and industry collaboration with W3C and OMG participation.

Several components for this work are already in place. We might be able to share and extend a common high level ontology, like the SOA ontology, offered by the Open Group [8], and follow the semantic service specifications by the Banking Industry Architecture Network (BIAN) [9].

Conversational scripts will be different for each case and with growing semantic support will transition to more sophisticated conversational and business scenarios with multiple branches to build on-the-fly decision trees.

The same approach and technology will support multiple cases that help the transition from the world of structured and unstructured data to linked and active knowledge and further to knowledge-driven architecture [10].

Ontology Modules and Ontology Integration; Links to Common Sense Ontology and Industry Standards

In the real world, everything is connected to everything else. But creating and using such a “universal” model would be counterproductive for solving specific problems. Assembling a specific model for every specific task would be prohibitively expensive unless such assembly can be easily done with a limited number of ready-to-go ontology modules. The “easily” part of integration is still a work in progress [11].

Multiple ontology modules represent multiple knowledge domains and views. 

Integration of modules and resolving possible conflicts can be done via ontology-maps.

It is important to be aware of and use existing “standard” ontologies. For example, the most comprehensive Common Sense ontology is available from Cycorp, Inc. [12]. The Common Sense ontology is necessary to understand and “conceptualize” unstructured text information, like the conversational entries of a Business Analyst, or like business documents.

I put the word “standard” in quotes. This is also a work in progress. For example, the Enterprise Data Management Council (EDMC) leads a collaboration of several financial institutions to create a semantic model of financial services. EDMC is teaming up with the Open Management Group (OMG) to turn the model into an upcoming standard ontology repository for the financial industry [13]. This standard will serve as a common ontological umbrella for financial firms and will provide basic reference data to regulators for systemic risk tracking purposes.

Summary

I have to admit that I’ve simplified several things. As Albert Einstein said, “Make everything as simple as possible, but not simpler.” And then he wrote his “Unified Field Theory” and everything that was written later on this subject was even more complicated.

One thing is clear: with the volume of information doubling every year, and with increasingly interconnected departments and corporations, old technology is on its way down (this could be quite a long way). Semantic technology is on its way up.

In the future, a new class of Knowledge Engineering will be introduced in every school along with the subject of Critical Thinking. Modeling tools that have Business and Development views today will add an Ontology view tab to the front page. This is happening as you read these lines.

The conversational approach is already adapted in its simplest form by many existing tools. Semantic support of the conversation is a very natural and powerful step ahead. We might be able to close one very boring chapter of the IT story, gradually decrease infrastructure expenses, and open the door for new exciting opportunities.

We will turn a beautiful idea of collaborative decision making into a working system [14] accessible by us, humans, and them, computer systems and services.

We will develop adaptive systems that can learn by conversing with people and store new skills as orchestrations of services [15]. These systems will manage complicated medication regimens for elderly people, call for help should an accident occur, participate in rescue operations and look for people trapped in a disaster, provide terrain mapping and manage operations in environments not suited for human beings. And we will definitely teach them to do IT maintenance work.


References:

1.      W3C on Large Triple Stores, http://www.w3.org/wiki/LargeTripleStores

2.      Realizing Efficiency & Interoperability: SOA & Semantic Technology in the Business Mission Area (BMA), U.S. DoD, Dennis E. Wisnosky, CTO, DoD (See Mr. Wisnosky’s 2011 Semantic Technology Keynote here)

3.      Integration-Ready Architecture and Design, Jeff (Yefim) Zhuk, Cambridge University Press, A book on Software and Knowledge Engineering

4.      Measuring IT Tactical Spending to Provide Transparency, Gartner Research

5.      Rules Collector, Yefim Zhuk/Boeing, From “tribal knowledge” to rules and rule-based applications

6.      World Wide Web Consortium, W3C, http://w3c.org

7.      RDF Services, http://www.w3.org/1999/11/02-RDFServices/

8.      Open Group Service-oriented architecture ontology, http://www.architecting-the-enterprise.com/blog/2011/01/the-open-group-service-oriented-architecture-ontology/

9.      Banking Industry Architecture Network (BIAN), http://www.bian.org/content/about_bian/index_en.html

10.      Knowledge-Driven Architecture, Yefim Zhuk, Streamlining development and driving applications with business rules & scenarios

11.      Open Ontology Repository (OOR) Initiative, http://ontolog.cim3.net/cgi-bin/wiki.pl?OpenOntologyRepository

12.      Cycorp, Inc. Knowledge Engineering and Common Sense Reasoning, http://cyc.com

13.      Enterprise Data Management Council (EDMC) and OMG, http://www.edmcouncil.org/PDFs/20110330.OMG.EDM.Relationship.pdf

14.      Adaptive Robot System with Knowledge-Driven Architecture, Yefim Zhuk, On-the-fly translations of situational requirements into adaptive robot skills

15.       Collaborative security and decision making in service-oriented environment, Yefim Zhuk/Boeing, Collaboration of multiple services and SMEs while processing many informational streams to optimize a strategy to achieve one or more goals.

 


Yefim "Jeff" ZhukDirector of Enterprise Architecture, Yefim leads Information Architecture at Sallie Mae. In the past he consulted government agencies and corporations in SOA and knowledge engineering, shared his expertise at Java One, Wireless One and Boeing Conferences. Cambridge University Press published his book “Integration-Ready Architecture and Design”. In the book and several patents he described a new field of Integrated Software and Knowledge Engineering and Knowledge-Driven Architecture. Hobby: mountaineering and guitar.  Yefim can be reached at: Yefim [dot] zhuk [at] salliemae.com.

1.      W3C on Large Triple Stores, http://www.w3.org/wiki/LargeTripleStores2.      Realizing Efficiency & Interoperability: SOA & Semantic Technology in the Business Mission Area (BMA), U.S. DoD, Dennis E. Wisnosky, CTO, DoD3.      Integration-Ready Architecture and Design, Jeff (Yefim) Zhuk, Cambridge University Press, A book on Software and Knowledge Engineering4.      Measuring IT Tactical Spending to Provide Transparency, Gartner Research5.      Rules Collector, Y. Zhuk/Boeing, From “tribal knowledge” to rules and rule-based applications6.      World Wide Web Consortium, W3C, http://w3c.org  7.      RDF Services, http://www.w3.org/1999/11/02-RDFServices/ 8.      Open Group Service-oriented architecture ontology, http://www.architecting-the-enterprise.com/blog/2011/01/the-open-group-service-oriented-architecture-ontology/ 9.      Banking Industry Architecture Network (BIAN), http://www.bian.org/content/about_bian/index_en.html 10.  Knowledge-Driven Architecture, Yefim Zhuk, Streamlining development and driving applications with business rules & scenarios11.  Open Ontology Repository (OOR) Initiative, http://ontolog.cim3.net/cgi-bin/wiki.pl?OpenOntologyRepository12.  Cycorp, Inc. Knowledge Engineering and Common Sense Reasoning, http://cyc.com13.  Enterprise Data Management Council (EDMC) and OMG, http://www.edmcouncil.org/PDFs/20110330.OMG.EDM.Relationship.pdf14.  Adaptive Robot System with Knowledge-Driven Architecture, Yefim Zhuk, On-the-fly translations of situational requirements into adaptive robot skills15.   Collaborative security and decision making in service-oriented environment, Yefim Zhuk/Boeing, Collaboration of multiple services and SMEs while processing many informational streams to optimize a strategy to achieve one or more goals