An overview of the ontology-derived Enterprise Data Model for retail and commercial banks.

The presentation was derived from a custom overview of the FIBO data model for a European retail and commercial bank. While every bank and data architecture team is different, challenges and solutions apply to most banks of similar size and semantic technology sophistication.

You can read the presentation or download the PowerPoint here.

Transcript of the lecture

Hi, this is Jurgen with the FIBO Data Model training course.
Today’s class is “Semantics for Midsize Banks,” but not only for banks. It’s fully applicable for Investment Funds. The licensing segment exceeds 200 billion US dollars in size. However, what it really means is banks that do not yet have ontologists and RDF stores.

Some background about the EDM Council and Object Management Group. The EDM Council is a global association of over 200 Financial Institutions. They focus on data management and data standards. In particular, many banks have completed the Data Management Capability Assessment Model. And for FIB-DM, it is important that the EDM Council created the FIBO.

The Object Management Group has over 220 member organizations.
They do standards like UML, BPML, DDS, and SysML, and they published the Commons Ontology Library. And now, as of October 1st last year, the two have merged. The EDM Council acquired OMG, and with that, the combined standard-setters are the world’s largest nonprofit trade organization for data management.

The FIBO is the authoritative model of financial industry concepts, definitions, and relations.
EDM council members, mainly the large banks and investment companies, developed the financial industry business ontology as a business conceptual model.
There are almost 2,500 classes that detail Loans, Financial Instruments, Business Entities, and Processes. Notably, the FIBO imports OMG Commons as an upper ontology.
OMG Commons is industry-independent. It holds generic constructs, whereas the FIBO is finance-specific.
The Financial Industry Business Data Model, FIB-DM, has almost 3,200 entities and includes the complete FIBO semantics and documentation.
And we created FIB-DM, using our patented Configurable Ontology to Data model Transformation (CODT).

There’s a chasm between semantic and conventional data management.
Large financial institutions have already implemented the FIBO. However, many midsize banks face high barriers: The tooling, specialized database systems, and the needed human expertise.
The EDM Council specified the FIBO in Ontology Web Language (OWL). However, OWL needs highly specialized ontologists. OWL runs on specialized databases, so-called RDF Triple Stores.
But IT departments must still support and design relational databases.

FIB-DM is the bridge across the chasm.
We have the industry standard available in our data modeling tool.
The FIBO in PowerDesigner, ERWin, and other tools.

Semantic Enterprise Information Architecture:
We can look at the IT landscape by its Use, whether it’s for Business, Design, or Development. The Type is Conceptual, Logical, Physical, or Implementation level?
Whether it’s for the Enterprise, Department, or Projects.
What you have today already: You likely have data models, RDBMS, and XSD for messaging. You may have Process Models, and you probably have Object Models in UML.
Now, what’s new with the FIBO is that the FIBO deploys on RDF databases, and with FIB-DM, we have the FIBO available as a Conceptual Data Model. And in the future, we can also generate Class Models and XSDs.
We have the FIP-CM and the FIB concept maps, which are in sync and aligned with both the data model and the ontology.

The Domain Ontology generates a perfect Conceptual Data Model.
On the left-hand side, we have a snippet of the FIBO. And what we see here is a Bank Account with a Bank Account Identifier.
We have the Bank, which is a Depository Institution.
And on the right-hand side, we see the data model, and it is the best representation of the left-hand side ontology graph.
There are no extra entities or missing entities and relationships in the design.

Financial Industry Business can be described with 15 Fundamental Concepts.
And in FIB-CM, the Concept Maps all have a name, an icon, and an abbreviation.
90% of FIB-DM entities are subtypes of Concept Entities.
In the ontology, these 15 classes, most of which are defined in OMG Commons, are only subclasses of the Thing. The other 2500 classes are um subclasses of a Concept Class. And that’s enough to visualize design in user-friendly Concept Maps.
If you want to dive deeper into that, I recommend the class “Semantics for Finance Users.”
FIB-DM comes with documentation. For each concept, there’s a reference sheet.
It has the FIBO definition. It has the number of entities under a concept, important subtypes, and it has the significant concept associations.

Here’s an example: the Agent. We see that the Agent entity in FIB_DM is broken down into Party, Persons, and Organizations.
And we also see how that concept relates to other concepts.
An Agent has an Address, which is a Reference.
The Concepts and Vocabulary directly correspond to the ontology.
On the left-hand side, we see a concept map.
It’s about the registration of depository institutions. In the US, they all have an FDIC Certificate Number that is registered with the FDIC and listed in a Registry.
The Depository Institution, our bank, is a Stock Corporation. It has an address, Capital, and an LEI.
This directly corresponds to the ontology. What we see here is a Protégé screenshot. If you want to dive into this deeper into this, I recommend “Scoping our first data model from FIB-DM.”

The data model and ontology are the same.
The data model ontology graph models the same business concepts and relations. Now, here’s an example. We look at the ontology. We have a Loan, and we can navigate to its Credit Arrangement via the Loan.
From there, the Principal, Payment Terms, and we find the Total Outstanding Principal. This is the identical representation scope from the FIBO Data Model.
CODT is the only technology that harvests the complete ontology semantics, including class restrictions and inverse object properties.
If you want to learn more about navigating between entities in the data model, I recommend the FIBO/FIB-DM Navigator class.

The Configurable Ontology to Data model Transformation (CODT).
On the left-hand side, we have ontologies such as FIBO, other industry ontologies, and our proprietary ontologies; on the right-hand side, we have FIB-DM, Enterprise, and Project Models.
We generate data models from our industry domain and proprietary ontologies, and we can also reverse-engineer our Data Models into RDF/OWL to extend the Enterprise and Project Ontologies.
The United States Patent and Trademark Office has issued patent #12038939 for CODT.
Yeah, a little excursus, the FIBO in ERWin.
FIB-DM is released as an SAP PowerDesigner Conceptual Data Model. However, ERWin has very good “import from external format” functionality, which makes migration to the ERWin modeling tool easy.
Here we see some statistics for the imported FIBO PowerDesigner file.
The input is complete, accurate, and fully preserves the extensive FIBO documentation in ERWin User-Defined Properties.
If you want to look into the import in more detail, you can do that on the FIB-UM website. There’s also a discussion on importing the FIBO PowerDesigner file into other data modeling tools.

Let’s look at the Data Model Packages.
At the top, we have a generic layer. There is OMG Commons, OMG languages, Countries, and Currencies. And some supplemental ontologies are used for documentation, like SKOS (Simple Knowledge Organization System).
Then here comes, in green, the domain core um that is FIBO Foundation, FIBO Business Entities, and Finance Business and Commerce.
Then come the domain extensions. For banks, of course, that’s Loans and Mortgages, and for uh Investment managers, that Securities, Derivatives, and Corporate Actions (CIV), in other words, Funds.
In the upper part, that’s the FIB-DM open-source version, and the green-shaded packages are in the commercial version.

The pass to SEIA:
This year, adopt the industry standard, the FIBO and its Data and Concept Map model. Then progress to training data architects in OWL and customize the FIBO to become your Enterprise Ontology.
With that, you’ll achieve Semantic Enterprise Information Architecture.
The benefit of this path is that we have an immediate Return on Investment from the FIBO content and design patterns from day one, not in two years.
And it’s easy to learn a new language like Ontology Web Language and its tooling if you already know the model’s content and design.

Version, support, and maintenance.
FIB-DM follows the quarterly EDM Council FIBO Production release schedule.
Licencees with a maintenance contract get the FIB-DM version four weeks after the FIBO release.
The current FIB-DM version is as of FIBO Q4 2025 and was released in January of this year.
Here we see the open-source and commercial versions side by side.
So, open-source is derived from 2018 Q4, and the Commercial Version is from 2025/Q4.
The open-source model has a little over 1,000 entities, and the commercial version adds 2,000 additional entities.
The open-source version is copyleft, whereas the commercial version is proprietary.
In other words, you keep your FIB-DM extensions, modifications, migrations, and scopes private.
Yeah, with the open source, the educational materials are under copyright, whereas with the commercial license, you’re free to leverage the original PowerPoints.

Further reading and watching:
The FIB-DM website has scalable vector graphics diagrams of the data model.
It has a list report of the 3,173 entities, complete with their definitions.
And extensive education materials, like this deck, in MS Office online and as a PDF download.
In the Education Course, we are here: Semantics for Midsize Banks.
You can also take a look at the Semantics for Large and Extra Large Banks.
They are all flavors of the same thing. The increasing size just means that the Excel bank presentation includes information about CODT and the Transformation Technology.
And from here, we go on to the Data Modeler classes: Semantics for Data Architects, Scoping, and the Navigator tool.
The website also has several articles with detailed descriptions. You can follow the FIB-DM LinkedIn page for news and updates, and watch this video and other educational videos on the FIB-DM YouTube channel.
Okay, thank you for watching