Chapter 13: Concepts and Language

1 The Ontological Base

Definition. It refers to a structured representation of knowledge in a domain. Describes the

entities, relationships, attributes, instances, and rules that govern a particular domain. It enables
categorization and reasoning, defining “what exists” and how domain components relate.

Entities (Concepts / Classes): Abstract representations of objects or ideas (e.g., ‘Person’,

‘Vehicle’).

Relationships: Define connections between entities (e.g., ‘owns’, ‘partOf’).

Attributes: Properties or characteristics of entities (e.g., ‘Age’ for ‘Person’).

Instances: Specific examples of entities (e.g., ‘Alice’ as an instance of ‘Person’).

Rules (Axioms): Logical conditions governing relationships (e.g., ‘A person can own at most
one driving license’).

Example 1 (Geographical Mapping). The ontological base organizes geographical entities to

connect them logically. This includes:

Entities: Countries, cities, mountains, rivers, etc.

Attributes: Elevation, location, or climate, etc.

Relationships: “City A is located in Country B”, “River C flows through City D”, etc.

Instances: Ganges River, Brahmaputra River, Prayagraj, etc.

Rules (Axioms): “ A river must flow downhill due to gravity.”, “A river must have a source
(spring, glacier, or lake) and a mouth (ocean, lake, or another river).”, etc.

For instance, the Ganges River has attributes like “length = 2,525 km” and “originates from
the Himalayas, while the relationship “flows through Northern India” links it to geographical
entities like cities and states.

1
Let’s look at the following graph:

What if we include the Brahmaputra River?

Example 2 (Business Organizational Structure). In business, the ontological base is used for
defining roles, hierarchies, workflows, etc. This includes:

Entities: Employees, departments, projects, etc.

Attributes: Employee roles, department budgets, project deadlines, etc.

Relationships: “Employee X works in Department Y”, “Project Z is managed by Employee X”,

Centre for Data Science, ITER, SOADU

etc.

Instances: Manager, Sales Department, etc.

Rules: “Each department must have at least one employee”. “Budget allocations must not
exceed the department’s financial limit”, etc.
For instance, an employee with attributes like “role = manager” and “salary = |75,000”,
with a relationship “works in the sales department” links him/her to “Project P handled by the
sales team.”
Let’s look at the corresponding knowledge graph:

Example 3 (Traffic Management System). The ontological base categorizes infrastructure and
vehicles. This includes:

Entities: Roads, traffic signals, vehicles, etc.

Attributes: Vehicle speed, road capacity, signal timing, etc.

Relationships: “Signal A regulates traffic flow on Road B”, “Car C drives on Road D”, etc.

Instances: Highway 101, Traffic Signal X, Car Model Y, etc.

Rules: “A vehicle must obey the speed limit assigned to the road.”, “A traffic signal must
regulate vehicles based on predefined timing cycles.”, etc.
For instance, a traffic signal has the attributes “green light duration = 30 seconds,” and a
relationship like “monitors Road R, which accommodates 10,000 vehicles daily.”
Let’s look at the corresponding knowledge graph:

Centre for Data Science, ITER, SOADU

2 Reification

Definition. Reification transforms abstract concepts, relationships, or ideas into concrete,

machine-readable, or analyzable entities. This allows for more detailed representation and ma-
nipulation of relationships by turning them into independent structured components.

Example 4. Consider the statement: “Rahul plays football.”

Non-reified version (or RDF):

Or, simply one can write: (Rahul, plays, football).

What if you wish to add extra details, such as who said, when said, and how true it is? Let us
assume that Coach said this on 20-05-2024 with a confidence level of 0.85. Then

Centre for Data Science, ITER, SOADU

 Reified version:

or,

(statement1, subject, Rahul)

(statement1, predicate, plays)

(statement1, object, football)

(statement1, source, Coach)

(statement1, date, 10-06-2015)

(statement1, confidence, 0.85)

Centre for Data Science, ITER, SOADU

Example 5. How can the statement “John is married to Alice” be represented using RDF?
Additionally, how can reification be used to include metadata such as marriage date (10-06-
2015) and place (New York)?

Ans:
RDF:

Or, (John, marriedTo, Alice)

Reified Version:

Centre for Data Science, ITER, SOADU

Or,

(Marriage123, subject, John)

(Marriage123, predicate, marriedTo)

(Marriage123, object, Alice)

(Marriage123, date, 10-06-2015)

(Marriage123, location, New York)

Example 6. How can the statement “Alice booked FlightXYZ” be represented using RDF, and
how can reification be used to include additional metadata such as the departure time (“2025-
04-01T08:00:00Z”) and seat number (“23A”)? Provide both the basic non-reified triple and a
reified version that models the booking as an event with detailed properties.

Example 7. How can the statement “Loan application was denied by the Bank” be represented
using RDF, and how can reification be used to capture additional information such as the reason
for denial (“Credit score < 600”) and the rule that was triggered (“PolicyRule789”)? Provide
both the basic non-reified triple and a reified version that models the decision as an event with
detailed properties explaining the outcome.

Note:

• When to Reify?
If some metadata is required to be attached.

• When Not to Reify?

If the facts are simple (without contextual data).

3 RDF & The Semantic Web

3.1 Resource Description Framework (RDF)

Definition. RDF is a framework for representing structured data using triples: Subject, Pred-
icate, and Object. RDF enables interoperability and machine-readability of information by
defining relationships and attributes in a linked data format.

Example 8. A patient (Patient123) is reported to have a fever and cough. It is known that any
flu typically has symptoms of fever and cough. Use RDF to represent this information. Based
on the RDF representation, what conclusion would you suggest?

Centre for Data Science, ITER, SOADU

Ans: Here, the RDF triples are:

• (Patient123, has, fever)

• (Patient123, has, cough)

• (flu, hasSymptom, fever)

• (flu, hasSymptom, cough)

Conclusion in RDF: (Patient123, has, flu).

Example 9. A user (User1) likes Movie1, which is of SciFi genre. Movie2, Movie3, and Movie4
are respectively in the SciFi, Action, and Comedy genres. Write the RDF triples that represent
this information. Based on the RDF representation, which movie can be recommended to the
user?

Ans: Here, the RDF triples are:

• (User1, likes, Movie1)

• (Movie1, hasGenre, SciFi)

• (Movie2, hasGenre, SciFi)

• (Movie3, hasGenre, Action)

• (Movie4, hasGenre, Comedy)

Since User1 likes Movie1, and Movie1 has the genre SciFi, we can infer that User1 likes SciFi
movies. Therefore, Movie2 can be recommended to the user.

3.2 The Semantic Web

Definition. The Semantic Web extends the web by making information machine-readable and
contextually linked. It leverages structured data models (such as RDF and OWL) to enable
intelligent querying, reasoning, and data integration.

It is an extension of the current web, where information is not just readable by humans but also
understandable by machines.
Today, most web data are presented in a way that appears meaningful to us but is merely text

Centre for Data Science, ITER, SOADU

to machines. The Semantic Web aims to make data machine-processable by giving it meaning
(semantics) using standardized formats.
Think of it as a web of data where machines can: understand relationships (like ’author
of’, ’part of’, ’located in’, etc.), draw conclusions (inference), and connect scattered pieces of
information.

Example 10. Consider the following information:

(a) Isaac Newton was born in England.

(b) He made major contributions to Physics and Mathematics.

(c) He studied at the University of Cambridge.

(d) He discovered the Law of Gravity.

(e) He was a member of the Royal Society.

(f) The Law of Gravity is a part of Classical Mechanics.

Draw a semantic graph using nodes to represent entities (underlined words) and labeled arrows
(edges) to show relationships such as discovered, born in, studied at, contributed to, member
of, etc.

Ans:

Centre for Data Science, ITER, SOADU

Example 11. Consider the following information:

(a) Albert Einstein was born in Germany.

(b) He studied at Zurich University.

(c) He invented the Theory of Relativity.

(d) The Theory of Relativity is a part of Physics.

(e) Albert Einstein was awarded the Nobel Prize.

(f) The Nobel Prize is awarded in the field of Physics.

Draw a semantic graph using nodes to represent entities (underlined words), and labeled arrows
(edges) to show relationships such as invented, born in, studied at, won, is a part of, awarded
in, etc.

Ans:

4 Conclusion and some real-life applications

Together, Ontology, Reification, RDF, and the Semantic Web form the foundation of a smarter,
interconnected digital world, where machines not only store data, but also understand and reason
with it.

Centre for Data Science, ITER, SOADU

Applications:

(a) Ontology:

– Medical Ontology to model diseases, symptoms, and treatments.

– E-Commerce for organizing product categories and specifications.

(b) Reification:

– Access control and data permissions in different fields, like Cloud-Based Collaborative
Platforms, Healthcare Systems, Legal & Government Systems, etc.

– Temporal Reasoning in different fields, like Historical Knowledge Bases, Dynamic

Recommendations, AI Personal Assistants, etc.

(c) RDF:

– Google Knowledge Graph uses RDF-style data to connect facts.

– Library systems for describing books, authors, and subjects.

(d) Semantic Web:

– Healthcare systems integrating patient records and treatment plans.

– Smart travel assistants that combine weather, location, and reservation data.

Centre for Data Science, ITER, SOADU