Semantic Conceptual Models Hans-Georg Fill Co-sponsored by the - - PowerPoint PPT Presentation

semantic conceptual models
SMART_READER_LITE
LIVE PREVIEW

Semantic Conceptual Models Hans-Georg Fill Co-sponsored by the - - PowerPoint PPT Presentation

Workshop on Graphical Modeling Language Development July 3, 2012 Kgs. Lyngby, Denmark SeMFIS: A Tool for Managing Semantic Conceptual Models Hans-Georg Fill Co-sponsored by the Austrian Science Fund: Grant Number: J3028-N23 Agenda


slide-1
SLIDE 1

SeMFIS: A Tool for Managing Semantic Conceptual Models

Hans-Georg Fill

Workshop on Graphical Modeling Language Development July 3, 2012 – Kgs. Lyngby, Denmark

Co-sponsored by the Austrian Science Fund: Grant Number: J3028-N23

slide-2
SLIDE 2

Hans-Georg Fill 2012 http://www.dke.univie.ac.at

Agenda

  • Motivation
  • Foundations
  • Requirements for Handling Semantic Aspects
  • The SeMFIS Approach
  • Use Cases
  • Conclusion and Outlook
slide-3
SLIDE 3

Hans-Georg Fill 2012 http://www.dke.univie.ac.at

Motivation

  • Conceptual models today widely used in industry and academia
  • Many different types of conceptual modeling languages: see projects
  • n www.openmodels.at
  • Some parts of semantics of conceptual models typically not made

explicit, given in natural language

  • Algorithmic processing of conceptual models interesting for analyses,

simulations, configuration of information systems etc.

  • Solution: Map conceptual models to semantic schemata / ontologies

Challenges:  How to do this for arbitrary modeling languages?  How to provide an open accessible solution?

slide-4
SLIDE 4

Hans-Georg Fill 2012 http://www.dke.univie.ac.at

Components of Modeling Methods

(Karagiannis and Kühn, 2002)

Meta Model = Model of the Abstract Syntax of a Modeling Language

slide-5
SLIDE 5

Hans-Georg Fill 2012 http://www.dke.univie.ac.at

Semantic Aspects in Conceptual Models

  • Basic assumption: Conceptual Models primarily

intended for human understanding and communication

  • Semantics of labels typically only given in natural

language

  • Meta models as a semi-formal specification: Only

syntax and type semantics are formalized

  • Machine processing of models requires

some kind of formal semantics

  • Ontologies as shared semantic schemata

to enable processing

slide-6
SLIDE 6

Hans-Georg Fill 2012 http://www.dke.univie.ac.at

Example for a Meta Model and a Model

Contact Customer Notify Customer

Customer reached

Activity Subsequent Class

Relationclass

Meta Model Model

Type semantics Type semantics Type semantics Type semantics Type semantics

Write E-Mail

Customer not reached

Decision

slide-7
SLIDE 7

Hans-Georg Fill 2012 http://www.dke.univie.ac.at

Spectrum of “Ontologies”

weak semantics strong semantics Taxonomy Thesaurus Conceptual Model Logical Theory

Relational Model Schema ER Extended ER XTM RDF/S UML DAML+OIL,OWL Description Logic First Order Logic Modal Logic Is Sub-Classification of Has Narrower Meaning Than Is Subclass of

(Obrst, 2003)

slide-8
SLIDE 8

Hans-Georg Fill 2012 http://www.dke.univie.ac.at

Requirements for Handling Semantic Aspects

  • Approach should be applicable to arbitrary modeling

languages and different types of ontology languages

  • Ontologies should be exchangeable (re-use aspects)
  • Exchange should be based on standards, e.g. OWL,

XML

  • Approach should be easy to use, despite the

necessary formal definitions

  • Support for distributed modeling and semantic

annotation of models in a web-environment

  • Open for further development by the scientific

community

slide-9
SLIDE 9

Hans-Georg Fill 2012 http://www.dke.univie.ac.at

The SeMFIS Approach

Goals:

  • Provide an open platform for handling semantic

aspects in conceptual models

  • Applicable to arbitrary modeling languages
  • Establish a community dealing with these aspects

Includes: – Semantic conceptual model types – Algorithms and support tools – Open accessible web services – Support tools and services http://www.openmodels.at/web/semfis/

slide-10
SLIDE 10

Hans-Georg Fill 2012 http://www.dke.univie.ac.at

SeMFIS Meta Model

Semantic Annotation Model Type Frames Ontology Model Type OWL Ontology Model Type Term Model Type

slide-11
SLIDE 11

Hans-Georg Fill 2012 http://www.dke.univie.ac.at

Use of SeMFIS Model Types

Annotation Model Instance Frames Ontology Model Instance OWL Ontology Model Instance Conceptual Model Instances Thesaurus Model Instance

slide-12
SLIDE 12

Hans-Georg Fill 2012 http://www.dke.univie.ac.at

SeMFIS Algorithms, Web Services and Support Tools

Currently available algorithms:

  • Exchanging models in XML format
  • Exporting frames ontology models in Protégé frames format
  • Transferring concepts from conceptual models into ontology

models via annotation specifications

  • Obfuscating model information using OWL inheritance

Currently available web services:

  • Access the contents of models via web interfaces (XML / Rest)
  • Generation of various graphical formats of models

Support tools:

  • Protégé OWL Export plugin for SeMFIS incl. upload

to an openmodels.at repository

slide-13
SLIDE 13

Hans-Georg Fill 2012 http://www.dke.univie.ac.at

Implementation of SeMFIS on ADOxx

  • ADOxx meta modeling approach developed since

1995, originally for business process management

  • Part of the commercial ADOxx meta modeling

Platform provided by the University spin-off BOC AG

  • >1000 customers worldwide
  • Available for free for projects within the Open Models

Initiative on www.openmodels.at:

– Open use and access of the platform – Online (RDP) and offline model editors – Currently > 25 academic projects based on ADOxx on Open Models – Variety of free services for developers and users, e.g. wikis, forums, developer tools, etc.

slide-14
SLIDE 14

Hans-Georg Fill 2012 http://www.dke.univie.ac.at

FDMM Formalism for Describing ADOxx Meta Models and Models

Meta2 Model Meta Model Model

domain range card

Set theory, first order logic statements

Implementation

(Fill, Redmond, Karagiannis., 2012 presented at ICEIS 2012)

slide-15
SLIDE 15

Hans-Georg Fill 2012 http://www.dke.univie.ac.at

SeMFIS Architecture

SeMFIS Algorithmic Extensions Modeling Analysis Simulation Evaluation HTML Generation Import / Export Web Service Component CORE (Modeling subsystem (CORE))= Database User interaction (Windows) User interface Application components Repository SeMFIS Rest Services User interaction (Web)

ADOxx Platform Web Environment Protégé Platform

Protégé Desktop Application SeMFIS Protégé Plugin

slide-16
SLIDE 16

Hans-Georg Fill 2012 http://www.dke.univie.ac.at

SeMFIS Use Case: Risk-adjusted BP Simulation

SeMFIS on ADOxx Protégé Toolkit Jess Rule Engine

XML/ pprj

10 20 30 40 50 60 70 80 90 Frequency Class

Histogram

ADOscript JessTab link New ontology concepts Simulation Results

(Fill, 2012 – presented at ECIS 2012)

slide-17
SLIDE 17

Hans-Georg Fill 2012 http://www.dke.univie.ac.at

SeMFIS Use Case: Semantic based Abstraction / Obfuscation of Model Information

ET: 2 Min.

Enter new address in customer database Assign address for account notifications

Retrieve cross- selling

  • ffers?

Client counseling

  • n cross-selling
  • ffers

ET: 1 Min. ET: 10 Min.

Prob.: 30% Prob.: 70% ET: 2 Min.

Manual customer interaction; Customer data Assign address for account notifications IT inter- action? Customer interaction

ET: 1 Min. ET: 10 Min.

Prob.: 30% Prob.: 70%

Ontology Abstracted BP Model Original BP Model

Customer interaction Counseling activities Product counseling Investment counseling … Selling activities Product selling Cross-selling IT interaction Customer Management Customer Database Cross-selling application Data Customer Data Name …

Semantic Annotation

1 2 3

  • wl:subClassOf
  • wl:Class

Sequence Flow Activity Decision Attribute

(Fill, 2011 – presented at BIR 2011)

slide-18
SLIDE 18

Hans-Georg Fill 2012 http://www.dke.univie.ac.at

Conclusion and Outlook

  • SeMFIS as a concept and a tool to handle semantic

aspects of conceptual models

  • Realization by using semantic conceptual models

together with algorithms, web services and support tools Future work:

  • Further development of the web

functionalities

  • Provision of evolution and change handling

facilities

  • Additional practical evaluation in research and

industrial projects

slide-19
SLIDE 19

Hans-Georg Fill 2012 http://www.dke.univie.ac.at

Thank you for your attention!

slide-20
SLIDE 20

Hans-Georg Fill 2012 http://www.dke.univie.ac.at

References

  • Fill, Hans-Georg, Redmond Timothy, Karagiannis, Dimitris (2012): FDMM: A Formalism for

Describing ADOxx Meta Models and Models, in: Leszek Maciaszek, Alfredo Cuzzocrea and José Cordeiro: Proceedings of ICEIS 2012 – 14th International Conference on Enterprise Information Systems, Vol.3, p.133-144, SciTePress (nominated for Best Paper Award).

  • Fill, Hans-Georg (2012): An Approach for Analyzing the Effects of Risks on Business

Processes Using Semantic Annotations, European Conference on Information Systems (ECIS'2012), AIS.

  • Fill, Hans-Georg (2012): Using Obfuscating Transformations for Supporting the Sharing and

Analysis of Conceptual Models, in: Robra-Bissantz, S., Mattfeld, D. (2012): Multikonferenz Wirtschaftsinformatik 2012, GITO.

  • Fill, Hans-Georg (2011): Using Semantically Annotated Models for Supporting Business

Process Benchmarking, in: Grabis, J., Kirikova, M. (eds:) (2011): Perspectives in Business Informatics Research - 10th International Conference, BIR 2011, Riga, Springer, 29-43.

  • Fill, Hans-Georg (2011): On the Conceptualization of a Modeling Language for Semantic

Model Annotations, in: Salinesi, C., Pastor, O. (2011): Advanced Information Systems Engineering Workshops, CAiSE 2011, LNBIP Vol. 83, Part 2, 134-148, Springer.

  • Karagiannis, D. and H. Kuehn (2002). Metamodeling Platforms. Third International

Conference EC-Web 2002 – Dexa 2002. K. Bauknecht, A. Min Tjoa and G. Quirchmayr. Aix- en-Provence, France, Springer: 182.