Date: Fri, 29 Mar 2024 14:52:11 +0000 (UTC) Message-ID: <1087709890.8127.1711723931293@ip-10-10-7-29.ec2.internal> Subject: Exported From Confluence MIME-Version: 1.0 Content-Type: multipart/related; boundary="----=_Part_8126_103845762.1711723931291" ------=_Part_8126_103845762.1711723931291 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Content-Location: file:///C:/exported.html
Description: A semantic data model describes the = concepts that are important to an organization along with their meanings an= d relationships to other important concepts. It is something like an = authoritative Glossary of Terms along with diagrams or other ways= of showing the relationships between different data. A semantic= model has an emphasis on relationships and meaning and how the data relate= to the real world. Ideally, the highest level semantic data model fo= r an organization should be able to fit on one page, and provide the abilit= y to drill down for more detail.
A semantic data model is not an entity relationship diagram; it is= not a relational model; it is not a UML class diagram, although UML practi= tioners sometimes use class diagrams to illustrate data concepts.
Goals: Goals of creating a semantic data model ar= e:
to get agreement and clarity on the meanings of concepts that = are important to an organization or a business domain.
to identify the systems of record that contain these important= concepts.
to identify where there are different contextual definitions f= or the same concepts within the organization.
to be able to provide the same definitions to end users whethe= r they access data through client applications which are consuming standard= APIs or through reports from the enterprise data warehouse.
A semantic data model is used in building common understandin= g of things that are important to the organization in achieving its goals a= nd objectives. It literally provides a common vocabulary.
A semantic data model focuses on the nouns and how they integrate = with one another. It is very useful for building consensus and general unde= rstanding throughout an organization. Because it is a conceptual model, the= scope is high level, but it can be scoped to any business domain.= p>
Source: From the Wikipedia article on semantic da= ta models:
The need for semantic data models was= first recognized by the U.S. Air Force in the mid-1970s as a result of the= Integrated Computer-Aided Manufacturing (ICAM) Program. The objective of this program was to increas= e manufacturing productivity through the systematic application of computer= technology. The ICAM Program identified a need for better analysis and com= munication techniques for people involved in improving manufacturing produc= tivity. As a result, the ICAM Program developed a series of techniques know= n as the IDEF (ICAM Definition) Methods which included the following:[1]During the 1990s the application= of semantic modelling techniques resulted in the semantic data models of t= he second kind. An example of such is the semantic data model that is stand= ardised as ISO 15926-2 (2002), which is further develope= d into the semantic modelling language Gellish (2005). The= definition of the Gellish language is documented in the form of a semantic= data model. Gellish itself is a semantic modelling language, that can be u= sed to create other semantic models. Those semantic models can be stored in= Gellish Databases, being semantic databases.
- IDEF0 used to produ= ce a =E2=80=9Cfunction model=E2=80=9D which is a structured representation = of the activities or processes within the environment or system.
- IDEF1<= /span> used to produce an =E2=80=9Cinformation model=E2=80=9D whi= ch represents the structure and semantics of information within the environ= ment or system.
- IDEF1= X is a semantic data modeling technique. It is used to pro= duce a graphical information model which represents the structure and seman= tics of information within an environment or system. Use of this standard p= ermits the construction of semantic data models which may serve to support = the management of data as a resource, the integration of information system= s, and the building of computer databases.
- IDEF2<= /span> used to produce a =E2=80=9Cdynamics model=E2=80=9D which r= epresents the time varying behavioral characteristics of the environment or= system.
A semantic data model can be used = to serve many purposes. Some key objectives include:
Planning of Data Resources<= /em>: A preliminary data model can be used to provide an overall view of th= e data required to run an enterprise. The model can then be analyzed to ide= ntify and scope projects to build shared data resources.
Building of Shareable Datab= ases: A fully developed model can be used to define an application ind= ependent view of data which can be validated by users and then transformed = into a physical database design for any of the various DBMS technologies. I= n addition to generating databases which are consistent and shareable, deve= lopment costs can be drastically reduced through data modeling. Identifying= business data that needs to be consistent and shared between different uni= ts. Having a single source of truth for central data that is at the core of= various business units.
Evaluation of Vendor Softwa= re: Since a data model actually represents the infrastructure of an or= ganization, vendor software can be evaluated against a company=E2=80=99s da= ta model in order to identify possible inconsistencies between the infrastr= ucture implied by the software and the way the company actually does busine= ss. Help identify areas where the design of the integration with 3rd party = software should be done via loosely coupled APIs.
Integration of Existing Dat= abases: By defining the contents of existing databases with semantic d= ata models, an integrated data definition can be derived. With the proper t= echnology, the resulting conceptual schema can be used to control transacti= on processing in a distributed database environment. The U.S. Air Force Int= egrated Information Support System (I2S2) is an experimental development an= d demonstration of this type of technology applied to a heterogeneous DBMS = environment.
Skills:
Ability to understand technical aspects of data/information
Ability to understand the nouns that relate to business proces= ses and capabilities.
Ability to do data/information modeling.
Ability to understand what is core data (entities) and what ar= e important attributes to capture
Roles:
Information Architect
Enterprise Architect
CIO
Business users
Others?
Steps:
Identify the different elements of the solution
Identify how the elements relate to each other
Tools:
Can be as simple as a database and a web front end. The = important part is wide visibility.
Visio/PowerPoint
Lucidchart
(to be completed)
UW-IT Investment Planning Objects = and Definitions:
UW Financial System Glossary:
ITANA reference architecture for teaching and learning (RATL): https://spaces.at.internet2.edu/display/itana/Conceptual+data+model+v04=
Note: We need a more canonical example to point to. UW is ex= ploring using graph databases to illustrate the semantic data model. = This should be a big improvement on a flat glossary of terms or a one page = description.
Capability Maps: Semantic Data Models can be aligned and uti= lized in concert with Capability Maps to engage stakeholders.
Process Maps: Semantic Data Models can help explain how differe= nt processes are linked and what data are acted on and how.
Synonyms:
canonical data model
conceptual data model
Research (related links):
Note that these resources previously hosted at databaseanswers.org have = been moved to https://datamodels.databases.biz/ and as of January 2024 are a= vailable there only partially. Those resources are available, also pa= rtially, at The Internet Archive ("IA" links included below).
Architecture Methods > Semantic Data Models= p>
Want to help with this page? Please see the Method Contributor Guide.
Stewards for this page:
Leo Fernig, University of British Columbia
Paul Schurr, University of Washington
Other contributors:
Dana Miller, Miami University of Ohio
Bob Dein, Miami (OH) University
Troy Martin, BYU
David Roberts, University of Michigan Medical School
Scott Fullerton, University of Wisconsin Madison
Jose Cedeno, Oregon State University
Rupert Berk, University of Washington
Robert Dean