Patent Application
20070112718
This application claims priority under 35 U.S.C. § 119 to Chinese Application No. 200510109557.X filed Oct. 25, 2005, the entire text of which is specifically incorporated by reference herein.
With the widely adoption of model-driven approach in today's enterprises, business is being modeled as large amount of models residing in multi-layers (like strategy, operation, execution, implementation and etc.). But the semantics (meaning) within these models are still not well integrated and thus makes these models hard to understand, communicate, reuse and utilize.
Business semantics in models can be roughly divided into two categories:
1. Model-level semantics (hereinafter referred to as model semantics)
This level of semantics is usually embedded in the model meta-structure. Using business process model as a sample, each modeling element (e.g., activity, data object, control and etc.) and the relationship among multiple modeling elements (e.g., one activity will send/receive a data object) stands for a specific and clear meaning.
2. Domain-level semantics (hereinafter referred to as domain semantics)
This level of semantics is usually represented as domain knowledge and referenced in the word/phrases used by the business models. Each word used in the title, caption, or comment for the modeling elements/relationship stands for some specific and exact meaning in the domain semantics.
Currently, we have the following challenges in well manipulating and leveraging business semantics.
Firstly, different business modeling methods have different meta-model as a specification of the model semantics, and only these models that share the same meta-model can share a common base of semantics.
Secondly, there is no well-established standard to regulate and interpret the meaning embedded in the words and phrases within business models.
And more importantly, the model semantics and domain semantics, albeit important to business, are usually isolated when are actually divided into two isolated parts.
Some existing methods and approaches have been proposed and applied by different companies and organizations, implicitly or explicitly, to address the problem as identified above.
Unified Modeling Language (UML) is designed by OMG (Object Management Group) to specify, visualize and document models for software systems. With an objective to be a common modeling language for different kind of software elements, UML can be regarded as an effort to unify the model semantics. Meta Object Framework (MOF) is the specification in UML to support meta-model specification. All the models that conform to the same meta-model will share a common base of model-level semantics and thus be inter-operable. Yet the problem for UML lies in 1) it is still very hard to enforce all the modeling method to use a common language (like MOF) as its meta-model description language; 2) two models following different meta-model (even both are MOF based) still cannot share model-level semantics completely, it is still hard to establish relationship among these models.
Ontology is regarded by many as the best way to address the problem related to domain semantics. RDF(S) and DAML/OWL are defined by standard organizations to formally represent the concept and their relationships within a specific domain (such as Banking and Telecom). The major challenge, however, lies in the difficulty to have a common ontology acceptable to all related parties.
Both approaches described above, actually, don't take enough consideration for the business semantics isolation problem. That is, with today's approaches, models are either looked as a set of meta-structure without domain content (which needs to be understood by human being), or just a concept/relationship hierarchy without connection with the environment to which it applied. Yet when we need to have further understanding and utilizing of the models, through computer aided mechanisms, both levels of semantics are not just important, but actually indispensable. This problem has become a major obstacle for the further application of model-driven approach in well attacking business problems.
As a summary, today's problems in utilizing business semantics can be summarized as:
1. The model semantics and domain semantics are usually isolated in real business practices, making their integrated computation impossible.
2. The domain semantics within business models are ambiguous and there lacks a mechanism to share domain semantics among business models that follow different meta models.
An aspect of the present invention provides a method and apparatus to integrate model and domain semantics in business models, which can overcome the aforementioned drawbacks of the prior art. To this end, an exemplary method for integrating model and domain semantics in business models is provided. The method includes a business model inputting step for inputting the business model to be realized; a domain semantic locating step for locating the domain semantics of the modeling element of the business model within the domain ontology and output the corresponding domain model semantics; a model semantics transforming step for transforming the modeling element of the business model into business model semantics that are represented by model ontology; and a unified semantic model forming step for combining the aforesaid business model semantics and domain model semantics and then outputting the formed unified semantic model.
Another exemplary aspect of the invention is an apparatus for integrating model-level and domain-level semantics in business models. The apparatus includes a domain semantics locator configured to locate the domain semantics of the modeling element of the input business model within the domain ontology and output the corresponding domain model semantics; a model semantics transformer configured to transform the input modeling element of the business model into business model semantics that is represented by model ontology; and a unified semantic model builder configured to combine the aforesaid business model semantics and domain model semantics and then output the formed unified semantic model.
One point addressed is how to associate and combine the two kinds of semantics (domain-level and model-level) embedded in business models, given the fact that they may be created by different people following different meta-model and domain notations. Furthermore, a unique approach of using USM as a common base for both the model and domain semantics embedded in models is presented. In contrast to the common usage of ontology to represent concepts within a particular domain, the invention creatively uses it to represent the model semantics for a particular modeling method. The Unified Semantic Model (USM) uses concept and relationship as fundamental modeling elements and is organized in the form of Ontology. During the integration of model semantics and domain semantics, business models will be transformed into the representation of USM, then domain semantics within business models will be specified and located to the corresponding concepts/relationships in domain ontology, by annotating the words/phases that are used in the caption or comments of modeling elements. After that, the semantics within business models, no matter as model or domain, will be transformed into a unified semantic model, which can then be used to support further work like analysis, inference and so on.
To guarantee the quality of the generated USM, an inference engine will be used to validate the USM based on some constraints embedded in the domain and model ontology, and user-provided rules or policies.
The steps of the method provided by the invention herein can be automated by algorithms in software or assisted by tooling with graphical user interfaces.
The following advantages may be achieved:
Firstly, both domain and model semantics are captured and utilized, making full usage of existing models will create more meaningful business result and value.
Secondly, all the models in known modeling method and format are supported, without the need to enforce any strong preconditions for the modelers.
Finally, the integration of business semantics into Unified Semantic Model makes it possible to have multiple further usages, the business value is tremendous.
These and other features, aspects and advantages of the present invention will more fully understood when considered with respect to the following detailed description, the contents of the accompanying drawings where:
In addition, the output of the apparatus is Unified Semantic Model that has been verified by constraints embedded in model/domain ontology, and rules/policies provided by users.
The apparatus 20 shown in
Furthermore, the apparatus 20 may optionally comprises model normalizer 27 and model ontology generator 28 to support the processing procedure of the apparatus 20. The said model normalizer 27 can check and normalize vocabularies used in captions or comments within business model elements, based on vocabularies in domain ontology 261, and then the normalized business model is input into the said domain semantics locator 21 and model semantics transformer 22 for appropriate process. The model ontology generator 28 can generate model ontology representation based on metamodels in UML or XSD.
A simple and complete example to realize the method of the present invention using the apparatus 20 thereof will be described below.
The apparatus 20 of the present invention may choose to generate model ontology 262 by the model ontology generator 28 based on metamodel prior the step S31 in
Then the model ontology generator 28 generate model ontology 262 based on the metamodel for business process. A specific example of the model ontology 262 generated based on the metamodel in UML of
The generated model ontology 262 will be stored in the Ontology Repository 26.
The apparatus 20 of the present invention also prepares domain ontology 261 prior step S31. The domain ontology 261 captures domain semantics that are used in business models. It comes from industry standards or domain experts, and can be reused.
The following example shows a domain ontology for banking.
The apparatus 20 of the present invention may optionally normalize the business model to be processed prior step S31.
Based on text analysis technologies, vocabularies used in the business process will be analyzed and normalized based on vocabularies in domain ontology. Referring to
Referring back to
In step S32, normalized business model will be transformed into business model semantics represented by model ontology by model semantics transformer 22. The domain semantics locator 21 will traverse the generated business model semantics, and then look up and locate the domain concepts or relationships occurs in business model semantics within domain ontology 261. The domain semantics locator 21 and model semantics transformer 22 will take domain ontology, model ontology and normalized business model as inputs. The domain semantics locator 21 will locate words/phrases within captions/comments in the modeling elements of the business model to corresponding concepts/relationships in domain ontology 261. And the model semantics transformer 22 will transform business models into business model semantics. The specific process flow of step S32 is as shown in
In
In step S33 in
Here is the sample unified semantic model that is generated for the business process shown in
In
to say that each “Activity” element in a business process model should represent for an “Operation” concept in the banking domain ontology;
to say that each “Object” element in a business process model should represent for a “Document” concept in the banking domain ontology
These constraints can also be transformed into the semantics model as follows:
Then we can input the unified semantic model, containing model ontology, domain ontology, instances of model ontology (generated from normalized business operational models) and constraint ontology, to Inference Engine and check the consistency of the semantics model.
Here are two more complex rules that are provided by users.
Rule1: We should do “inquiry” before any kind of “transaction”
It is apparent that Rule 2 will make the semantics model generated in
The method according to the present invention may be encoded as program stored on the computer-readable storage medium, and can be realized by executing the program by a computer. Therefore, the present invention also includes the computer program product that is encoded according to the method of the present invention, as well as the computer readable medium storing the said computer program.
The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used herein, the singular forms “a”, “an” and “the” are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms “comprises” and/or “comprising,” when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
The corresponding structures, materials, acts, and equivalents of all means or step plus function elements in the claims below are intended to include any structure, material, or act for performing the function in combination with other claimed elements as specifically claimed. The description of the present invention has been presented for purposes of illustration and description, but is not intended to be exhaustive or limited to the invention in the form disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the invention. The embodiment was chosen and described in order to best explain the principles of the invention and the practical application, and to enable others of ordinary skill in the art to understand the invention for various embodiments with various modifications as are suited to the particular use contemplated.
Having thus described the invention of the present application in detail and by reference to embodiments thereof, it will be apparent that modifications and variations are possible without departing from the scope of the invention defined in the appended claims.
| Number | Date | Country | Kind |
|---|---|---|---|
| 200510109557.X | Oct 2005 | CN | national |
