DAMA + Wilshire Meta-Data Conference - Data Strategy Track
This presentation recounts work done at General Motors in the area of data pattern identification and use. GM has developed an extensive catalog of data patterns to facilitate work effort in integrating data across the GM enterprise. The patterns help drive semantic reasoning and determine information appropriateness for integration. In addition, the data patterns are also used to derive work effort and costing associated with information integration initiatives.
  •  The IT integration imperative
  •  Fact, fiction, and reality
  •  Semantic disintegrity propagation
  •  Data pattern examples
Jack Myers of the Metadata Company will present current cutting edge ontology integration R&D conducted for the USN to support critical infrastructure protection. The research focuses on issues associated with the integration and use of complex vocabulary-based ontologies in the workplace. Representative issues include resolving difficulties creating statements and queries in triples such that they are not unmanageably large; and, fully integrating ontologies so that not only will classes and relationships between them that are common to each integrated ontology be available, but for all of these to be available to each ontology. Jack will discuss the issues addressed by the research, the use of a simple semantic ontology data language used to conduct the research along with the results. Examples will be offered using the Unified Medical Library System (UMLS) and the Gene Ontology (GO).

Implications for the practical use of ontologies in the marketplace using the semantic ontology language will be presented Bullet Points:
  •  A detailed description of the structure of the ontology model that is being used (which will be in the public domain in 2007) will be provided along with examples of queries across multiple domains.
  •  Specific examples of how other ontologies can be mapped to the generalized model will be shown. Other medical ontologies (e.g., National Institute of Health’s Oncology Ontology, the OBO, and other Healthcare Ontologies and data models) are candidates to be added to this ‘knowledgebase’.
  •  The intermediate goal of this R&D project is to demonstrate that all medical and health knowledge can be fully integrated into one general model using the basis of the semantic ontology language. Previous work demonstrated that queries can be developed that can access needed data from externally defined sources to provide cross-domain solutions.
  •  Mr. Myers will show how existing healthcare data can be integrated into the ontologies.
Enterprise modelers are increasing interested in the semantic integrity and implementation-independence of the business models that underly service-oriented architectures and object or data models used in applications. The semantic web framework offers the most mature and broadly accepted semantic modeling capabilities and the most logically clean and capable formalisms for rules available. This framework include OWL, the web ontology language and the semantic web rule language (SWRL) and its first-order logic extension (SWRL-FOL).

This presentation demonstrates an upper ontology that provides a semantic core (e.g., general purpose units, time, the calendar, numbers, entities) for OWL ontologies and the linguistic augmentation of OWL ontologies such that human language (e.g., English) sentences can be formally understood and translated into SWRL-FOL for execution by a general purpose inference engine.

The demonstration is couched in a financial services application accessed through a web browser, the behavior of which is determined by sentences using a vocabulary defined using the rooted OWL ontology.

What is the look and feel of a typical business rules project? In this presentation, Mr. Ross uses a real-life case study from the insurance industry to illustrate in detail how structured business vocabularies play an essential role in the capture, expression and analysis of business rules.

Starting with source underwriter guidelines for auto insurance, Mr. Ross takes you step-by-step through business rule methodology. He highlights how the business vocabulary is built and portrayed, and is then used to identify gaps, discrepancies and anomalies in the rules. He explains what deliverables, techniques and tools you’ll need to support the effort, and how to ensure success both in your interactions with business people, and in moving toward implementation.
  •  The business rules approach to vocabulary and rules.
  •  Fact models – visualizing the business vocabulary.
  •  Fact models vs. data models.
  •  Reduction of rule logic.
  •  How to organize and analyze large, complex sets of rules.
  •  RuleSpeakR, structured business-friendly syntax for expressing business rules.
  •  Tips and tricks.
The hot new buzzwords in the data management world are actually taken from ancient Greek philosophy: Semantics and Ontology. Ontology asks the question "what actually exists?", and semantics is about how to describe it. Data modelers have been trying to capture the semantics of organizations for many years, but attempts to integrate systems that speak "different languages" have brought the topic to the forefront. Among other things, the Web Ontology Language (OWL) has been created to describe a company's ontology in a way that makes it possible to manipulate it with a computer. This paper will describe the language, its relationship to data modeling, and its strengths and weaknesses.
Topics included will be:
  •  A history of the language and its relationship to the Semantic Web.
  •  Its basic syntax.
  •  Its relationship to data modeling: the two techniques approach essentially the same problem from two very different directions.
  •  Using OWL to draw inferences from data.
  •  The limitations of OWL.
Service Oriented Architecture is very hot now: most survey have about 60% of large firms starting an SOA project this year. Left to their own devices, many SOA projects will go bad. One of the main reasons is that a purely developer lead project has a tendency to re-implement Distributed Objects with XML interfaces, instead of applying the discipline needed to get to truly reusable services and messages. In this session will we review a methodology we have used successfully on several engagements we call the "Enterprise Message Model" (EMM). In this session attendees will:
  •  Learn the basics of SOA - Be able to describe why letting an SOA project run open loop is likely to fail
  •  Understand the basics of the EMM methodology Data Modeling professionals are in a great position to assist their firms with their SOA projects, if they seize the opportunity.
Data Modeling professionals are in a great position to assist their firms with their SOA projects, if they seize the opportunity.

For a growing number of companies, the main thing that stands in the way of true knowledge management and unlocking the value from their data is the inconsistencies in semantics.

Business semantics is defined as a consensus of terms and definitions that characterize:
  •  The business you are in
  •  The products and services you deliver; and
  •  Organizational knowledge. Business semantics management is emerging as a core component of service oriented architecture (SOA), which distributes applications that perform services on demand. For many experts, SOA is the future of software, where applications, used internally or externally, are delivered as services over the Internet.
This presentation will provide a broad overview of business semantics management, semantics as a service and case studies from 3-4 organizations in the publishing and technology industries.

Resource Description Framework (RDF) is a W3C standard format for storing arbitrary data on the web and elsewhere. It's particularly good for storing metadata about files and other machine-accessible resources. You can store handfuls of RDF inside the resources they describe, outside of them, in relational databases, in XML, or any place you like, and then easily combine these handfuls into a database that you can use for queries, reports, and graphs.

The ability to exchange and combine RDF from different places across the Web has made it a cornerstone of the W3C's "Semantic Web" plans, and it's already proving itself very helpful in accomplishing much more mundane tasks. Whether your data is structured or unstructured, typed or untyped, centralized or distributed, RDF just may make the job of storing and using that data easier.
  •  a data model, not a syntax; several syntaxes available
  •  if you can name something with a URI, you can assign it RDF metadata
  •  Building applications that can use RDF
  •  well-known RDF metadata: Adobe's XMP, FOAF, Dublin Core
  •  building and using ontologies with RDF/OWL
  •  RDF query languages
An Enterprise Ontology is a Semantic Model of the key concepts of a firm. It resembles a conceptual model but goes much beyond a conceptual model in that the formal definitions allow the system to infer class membership based on properties. In this workshop we will show both how to build an ontology using OWL/DL and explain how such an ontology can reduce complexity in the architecture and applications based on it.

Participants will:
  •  Gain an understanding of what an ontology is, and how it differs from a conceptual data model.
  •  Learn the basics of OWL/DL the standard ontology language of the Semantic Web
  •  Step through live demos showing ontologies being built and used.

This presentation is fairly technical and fast paced, but it has been designed specifically for the data management professional, so analogies will be made to similar DM concepts where ever possible.



Close Window
Wishire Conferences DAMA International