Method and an apparatus for automatic semantic annotation of a process model

• US 20090234640A1
• Filed: 02/05/2009
• Published: 09/17/2009
• Est. Priority Date: 03/13/2008
• Status: Active Grant

• Current Assignee: Siemens AG
• Original Assignee: Siemens AG
• Inventors: Boegl, Andreas, Weber, Norbert, Schrefl, Michael, Pomberger, Gustav
• Granted Patent: US 8,650,022 B2
• Time in Patent Office: Days
• US Class Current: 704/9
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G06F 8/10 Requirements analysis; Specification techniques

Abstract

An apparatus and a method for automated semantic annotation of a process model having model elements named by natural language expressions, wherein said apparatus comprises at least one semantic pattern analyser which analyses the textual structure of each natural language expression on the basis of predefined semantic pattern descriptions to establish a semantic linkage between each model element to classes and instances of a reference process ontology for generating a semantically annotated process model.

Claims

1. A method for automatic semantic annotation of a process model which comprises model elements named by natural language expressions,

wherein a textual structure of each natural language expression is analyzed on the basis of predefined semantic pattern descriptions to establish a semantic linkage between each model element to classes and instances of a reference process ontology for generation of a semantically annotated process model.

2. The method according to claim 1,

wherein the reference process ontology is formed by a lexical knowledge base comprising a vocabulary used for model elements and a process knowledge base comprising process semantics of model elements.

3. The method according to claim 1,

wherein the process model is formed by an EPC (event process chain) model, an UML (unified model language) model or a Petri network.

4. The method according to claim 1,

wherein each semantic pattern description comprises a semantic pattern template of a model element,

a set of template structures, and

a set of analysis rules.

5. The method according to claim 4,

wherein a semantic pattern template is formed by a domain name, and

an ordered tuple of variables each pointing to a class of said process knowledge base within the reference process ontology and to an instance of said class.

6. The method according to claim 4,

wherein the template structure is formed by

a unique identifier, and

an ordered tuple of variables each pointing to a class of said process knowledge base within said reference process ontology and to an instance of said class.

7. The method according to claim 4,

wherein an analysis rule is formed by

a precondition which compares template structures with term structures extracted from natural language expressions, and

an operator that generates one or several instances of semantic pattern templates which are assigned to a model element of said process model.

8. The method according to claim 1,

wherein a named process element is a function provided for performing a task on a process object to transform said process object from an initial state to a resulting state, or an event comprising a text clause.

9. The method according to claim 1,

wherein for each named model element of said process model a linguistic normalization of its natural language expression is performed by a term normaliser to generate a semantic label of said model element.

10. The method according to claim 9,

wherein a semantic pattern analysis of each generated semantic label is performed by a semantic pattern analyser to generate an ordered set of tuples containing instances of semantic pattern templates by using analysis rules of a semantic pattern description.

11. The method according to claim 10,

wherein said instantiated semantic pattern templates are assigned to the respective model element to generate said semantically annotated process model.

12. The method according to claims 1 to 11,

wherein the method is performed by executing a computer program.

13. The method according to claim 12,

wherein the computer program is loaded from a data carrier storing said computer program.

14. A method for automatic semantic annotation of a process model having model elements named by natural language expressions comprising the steps of:

performing a linguistic normalization of the natural language expressions to generate a semantic label for each named model element of said process model; and

performing a semantic pattern analysis for each generated semantic label to establish a semantic linkage between each model element to classes and instances of a reference process ontology.

15. An apparatus for automated semantic annotation of a process model having model elements named by natural language expressions,

wherein said apparatus comprises at least one semantic pattern analyser which analyses the textual structure of each natural language expression on the basis of predefined semantic pattern descriptions to establish a semantic linkage between each model element to classes and instances of a reference process ontology for generating a semantically annotated process model.

16. The apparatus according to claim 15, further comprising a term normaliser which performs a linguistic normalization of said natural language expressions of said named model elements to generate a semantic label for each named model element of said process model.

17. The apparatus according to claim 16,

wherein said apparatus further comprises a memory for storing said reference process ontology consisting of a lexical knowledge base comprising a vocabulary used for model elements, and

a process knowledge base comprising process semantics of model elements.

18. The apparatus according to claim 15,

wherein said process model comprises an EPC (event process chain)-model, a UML (unified model language)-model or a Petri network.

19. An apparatus for automated semantic annotation of a process model having model elements named by natural language expressions comprising:

means for performing a linguistic normalization of a natural language expression to generate a semantic label of the corresponding model element; and

means for performing a semantic pattern analysis of said generated semantic label to generate an ordered set of tuples containing instances of semantic pattern templates using analysis rules of a semantic pattern description.