OpenMechanics.org    

Business Ontology as a Basis for Information Systems Design

Ivan Mikushin, OpenMechanics / Ural State Technical University

Business information systems (ERP, CRM, DMS etc.) use Database Management software for services like data storage, retrieval and transaction processing, and Middleware software for services like business logic execution, security management and distributed transaction processing. That’s because all of them store, process and provide access to business data.

Data is the most important resource in a today’s enterprise. And because there are so many different business information systems addressing different data processing needs, it is distributed across them and presented in many different forms. The more so in different organizations. And this is one of the reasons that makes Business Process Automation so difficult and costly: to be able to unify business processes one should first unify the business data.

This is largely addressed by a concept of web services: protocols like SOAP (Simple Object Access Protocol) the languages like WSDL (Web Services Description Language) and IBM’s WSFL (Web Services Flow Language).

But web services only unify communication among co-existing information systems. But in building new information systems developers still need to reinvent the wheel and design their own database schemas and other data structures internal to the system, but still having to address business needs.

Let’s look at the problem from another point of view. What kind of data is presented in an enterprise information system? What does it describe? Eventually all the data circulating in these information systems is about business of the enterprise. Some of it describes business processes and documents, some of it describes IT resources such as users and workstations — any other aspect of the company’s resources can be described. Different data formats are used because of the existing software covering different aspects of the company’s business. And every new piece of software adds its own.

All this information could be used in a uniform way. Just such a way of business data unification is Business Ontology.

The philosophical meaning of the word “ontology” is “the philosophical study dealing with the nature of being, reality, ultimate substance”. This is close to our subject because with information we try to describe what exists in our business. Another, more practical meaning is “a particular theory about being or reality”. It is closer to our definition which is “an explicit specification of conceptualization”. Conceptualization is a description of a set of objects and terms and their relationships. By describing the business domain in this way we are dealing with knowledge (as opposed by data).

Now there exist a broad range of different knowledge description languages. For our purpose we will look at RDF (Resource Description Framework). Being a standard endorsed by the World Wide Web Consortium RDF is widely known and this a very important because we need a language that could be wide used and understood. The broad goal of RDF is to define a mechanism for describing resources that makes no assumptions about a particular application domain, nor defines (a priori) the semantics of any application domain. The definition of the mechanism should be domain neutral, yet the mechanism should be suitable for describing information about any domain.

Any object of interest can be identified by a URI and we could describe any relationships (also identified by URIs) between objects. These relationships are expresses as triples . Information about the business domain in this form could be easily used and manipulated by business information systems.

Thus we need a an enterprise wide ontology through which applications could access business objects: it could be viewed as a global catalog of business objects. And this catalogue doesn’t need to store all the data. Though lots of information processing needs might be already addressed by existing applications and objects represented by their data might be needed by our new application, we could write adapters to this data for our ontology server (the catalogue) and then transparently use these objects as though they are directly accessible.

Looking at business data from the ontological point of view we can see the following approach to building business information systems:

References

  1. T.R. Gruber. A translation approach to portable ontologies // Knowledge Acquisition. No. 5(2). – 1993.

  2. N. Guarino. Ontologies: What Are They, and Where’s The Research? –1996.

  3. Resource Description Framework (RDF) Model and Syntax Specification. – World Wide Web Consortium, 1999.

  4. Web Services Description Language (WSDL) 1.1. – World Wide Web Consortium, 2001.

  5. Web Services Flow Language. – IBM, May 2001.



© 2002/2010 OpenMechanics.org
SourceForge Logo