CEN/WS XBRL
CWA1:
•DPM Meta model
10 December, 2013
Katrin Heinze, Bundesbank
Page 1
CWA1
DPM representation
Definition of a Data Point Model




DPM is a dictionary of business concepts and their properties
used in tables (explicitly indicated in annotation)
identifying the content of every data point and
its relation to other data points.
Developed by:
Data Point Model
EIOPA
EBA
Identified obstacles with DPMs
Background for the development
 ambiguity in the understanding of the concept of a Data
Point Model
 missing rule set to be followed in the process of Data
Point modelling
 different constraints depending on the field of
application
 missing abstraction layer to ease the understanding for
IT experts
 missing description of the relations between the
different components of a Data Point Model
Objectives: DPM Meta model
Objectives
 represent the requirements of European reporting frameworks in a
data model
 link the business requirements with the technical transfer format
 describes the rules to be recognised in the process of modelling
The Data Point Meta Model should provide
 (1) the model components for the creation of a formal model on
sets of data points for European supervisory reporting frameworks,
 (2) rules on how to combine these components and
 (3) the meaning (semantic) of the components and their relations.
 Similar to a model construction kit for
toys it provides the modelling principles
with all characteristics available for use
by the modeller.
Multidimensional terminology
Total
Domain
List of Countries
European countries
France
Spain
Austria
Hierarchy
Belgium
Germany
Italy
Other than European countries
Japan
USA
5
Dimension
Multidimensional terminology samples
Domain
A Domain is a classification system to categorize items that share a common
semantic identity. A Domain provides therefore an unambiguous collection of
items in a value range. The items of a Domain can have a definite, and
therefore countable, number of items, or an infinite number of elements that
follow a specific (syntax) pattern.
Dimension
A Dimension is a data set to one characteristic area which is composed of
individual and non-overlapping data elements. In the context of a data point
model Dimensions are used to group information in a meaningful way.
Dimensions are used to define "by" conditions and provide structured
information to describe a data point in detail.
6
Reality
Metaphor
Concept of modelling
Meta model
modelling approach
model representation
Objekt system
Model system
modelling
objectives
7
Modeller
Process of DPM modelling
Levels of meta modelling
Meta meta
model
defines notation
conforms to
Meta model
conforms to
represented by
9
Data Point Model
Meta
modelling
language
conforms to
defines notation
Modelling
language
UML Meta model for DPM
Decisions taken on the
modelling approach for the
DPM Meta model
UML Model
defines notation
conforms to
Data Point
Meta model
UML
conforms to
defines notation
UML + OCL
conforms to
represented by
Data Point Model
OCL (Object Constraint Language) is a
formal language for specifications. It refers
to an UML model to describe constraints
about the objects in the model.
Perspectives of a meta model
Different perspectives on the meta model
 Perspectives reduce the complexity and provide views on different
aspects of the object system
 Perspectives needed to be consistent and complete as a whole
Front view
Side view
modelling
Top view
Inconsistency
Object system
Model system
DPM model perspectives
Perspectives on the DPM Meta Model
 Structural
 Versioning
 Dimension Validation
 Hierarchical
 Presentation
Constraints on the DPM modelling
Definition of constraints



General constraints on DPMs
Data warehouse specific constraints
European XBRL Taxonomy specific constraints
UML Meta model for DPM
The DPM Meta Model
 eases the understanding of DPMs for IT experts by using the
standard modelling language UML,
 reduces the complexity of DPMs by showing only the relevant
aspects,
 provides syntax and semantics to ease the automation of IT
tasks
 like generating data formats for the reporting process or
 validation checks on basis of the constraints defined,
 enables the derivation of a database design (relational as well
as multidimensional).
Model-2-Model transformation
Explanation of the correspondence of model graphs via
graph transformation by using the Triple Graph Grammar
 Here: mapping between DPM and XBRL
Domain graph
XBRL
correspondence graph
Domain
graph (DPM)
15
Model-2-Model transformation
Future fields of application
 Here: mapping between DPM and SDMX
«metaclass»
Domain
«metaclass»
:Attribute
-code
1
«metaclass»
:Attribute
-label
«metaclass»
:Attribute
A2A
-id
A2A
«metaclass»
:Attribute
-name
«metaclass»
Codelist
«metaclass»
Hierarchy
C2C
*
«metaclass»
:Attribute
-validFrom
1
1..*
«metaclass»
HierarchyRelationship
-parentCode
-childCode
Domain
graph (DPM)
1
«metaclass»
:Attribute
-validTo
1
Domain graph
(SDMX)
«metaclass»
:Attribute
-isPartial
*
2
«metaclass»
DefinedMember
1
C2C
correspondence graphs
«metaclass»
Code
parent
hierarchy
0..*
16
child
Further CWA1 deliverables
European data
point
methodology
European
reporting
frameworks
Guidelines for
data point
modelling
European XBRL
Taxonomy
Architecture
XBRL taxonomy
Data Point
Model
represented by
transformed by
XBRL instances
European
Filing Rules
17
Thanks for your attention
katrin.heinze@bundesbank.de
Comments or questions?
Page 18
CWA1