Introduction to RDF

Why RDF?

• Resource Description Framework (RDF) provides a standardized universal model for

representing data and its meaning
• Support hybrid, varied, and changing data models with ease
• Easy to represent any change in data or schema
• Interoperable and composable
• Eg. The song with the name “Love Me Do” has two
:Paul_McCartney
:Paul_McCartney
writers, Paul McCartney and John Lennon
“Love Me Do”

:wr
ite

e
am
r

:n
:writer
:John_Lennon :Love_Me_Do
Key Terms 1: The Basic Idea

• Let’s say the class (ie. category) “artist” includes both solo artists and
bands, and a member of a band is a solo artist
• The RDF way to describe these relationships is based on how we
would express it in speech:
Predicate
Subject Object

Eg. The Beatles has as a member Paul McCartney

:member
:The_Beatles :Paul_McCartney
Key Terms 2: Objects

• By class we mean a type of thing (eg. band or artist)

• A class is made up of a set of individuals (eg. The Beatles or John
Lennon), which can also be called instances or objects
• A class or individual can be the subject or the object in a 3-part RDF
structure called an RDF triple
Key Terms 3: Properties

• The middle part of an RDF triple is the predicate, which is used in two
ways. When it describes a relationship between two objects (classes
or individuals) in our model, then it is called an object property

Subject Predicate Object

• If the predicate provides data (a number, date, string, etc.) about an

object, it is called a data property describing an attribute

Subject Predicate Data

Key Terms 4: Graphs
• Taken together, these
elements make up a
graph
• In a graph, points
representing objects or
data are called nodes
while the predicates
that connect them
(either object properties
or data properties) are
called edges
Key Terms: Review
• There are two kinds of objects: classes and individuals/instances
• Classes are sets, collections, types of objects, kinds of things
• Individuals (or instances) are what a class groups together
• Properties come in two types. An object property is a relationship between
two things. Datatype properties are attributes of one thing
• In graph representation diagrams, classes & individuals are called nodes while
properties are called edges
• In RDF triples, classes and individuals are the subjects or objects, while
properties correspond to predicates. A set of RDF triples is called an RDF
graph
RDF Concepts

• IRI: Nodes and edges with a unique identifier

• Literal: Nodes representing values like numbers and dates
• Blank node: Nodes without an explicit identifier
IRI
• Internationalized Resource Identifier

http://www.w3.org/1999/02/22-rdf-syntax-ns#type

http://stardog.com/tutorial/The_Beatles

mailto:[email protected]

urn:isbn:9788026874256

tag:stardog.com,2018:product:stardog
Prefixed Name

• An IRI looks like this

http://www.w3.org/1999/02/22-rdf-syntax-ns#type

• Using a prefix declaration for its namespace

PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>

• Can be shortened to a prefixed name

rdf:type
Literals
• Literals are written in quotes followed by their datatype IRI
“1963-03-22” ^^xsd:date “1963-03-22T21:44:00Z”^^xsd:dateTime

• Datatype can be omitted for strings:

“The Beatles” “The Beatles”^^xsd:string

• Datatype and quotes can be omitted for some datatypes

125 3.14 3.2E4 true

integer decimal double boolean

RDF Serialization

• Declare prefixes

PREFIX :<http://stardog.com/tutorial/>
PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>

• Write subject, predicate, object followed by a ‘.’

:The_Beatles rdf:type :Band .

:The_Beatles :name “The Beatles” .
Turtle Syntax

PREFIX :<http://stardog.com/tutorial/>
PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>
PREFIX xsd: <http://www.w3.org/2001/XMLSchema#>

:Love_Me_Do rdf:type :Song .

:Love_Me_Do :name “Love Me Do" .
:Love_Me_Do :length “125” ^^xsd:integer .
:Love_Me_Do :writer :John_Lennon .
:Love_Me_Do :writer :Paul_McCartney .
Literal Shorthand

PREFIX :<http://stardog.com/tutorial/>
PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>
PREFIX xsd: <http://www.w3.org/2001/XMLSchema#>

:Love_Me_Do rdf:type :Song .

:Love_Me_Do :name “Love Me Do" .
:Love_Me_Do :length 125 .
:Love_Me_Do :writer :John_Lennon .
:Love_Me_Do :writer :Paul_McCartney .
Shorthand for rdf:type

PREFIX :<http://stardog.com/tutorial/>
PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>
PREFIX xsd: <http://www.w3.org/2001/XMLSchema#>

:Love_Me_Do a :Song .
:Love_Me_Do :name “Love Me Do" .
:Love_Me_Do :length 125 .
:Love_Me_Do :writer :John_Lennon .
:Love_Me_Do :writer :Paul_McCartney .
Same Subject

PREFIX :<http://stardog.com/tutorial/>
PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>
PREFIX xsd: <http://www.w3.org/2001/XMLSchema#>

:Love_Me_Do a :Song ;
:name “Love Me Do" ;
:length 125 ;
:writer :John_Lennon ;
:writer :Paul_McCartney .
Same Subject and Predicate

PREFIX :<http://stardog.com/tutorial/>
PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>
PREFIX xsd: <http://www.w3.org/2001/XMLSchema#>

:Love_Me_Do a :Song ;
:name “Love Me Do" ;
:length 125 ;
:writer :John_Lennon ,
:Paul_McCartney .
Ignore Whitespace

PREFIX :<http://stardog.com/tutorial/>
PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>
PREFIX xsd: <http://www.w3.org/2001/XMLSchema#>
:Love_Me_Do a :Song ;
:name “Love Me Do" ;
:length 125 ;
:writer :John_Lennon , :Paul_McCartney .
Blank Nodes
SPARQL
Triple Patterns

• A triple pattern is a triple with zero or more variables

?band rdf:type :Band . ?album :artist ?artist .

:Love_Me_Do :writer ?writer . :The_Beatles ?p ?o .

• Triple patterns match the triples in the graph

• Each matching triple produces one result
SELECT Query: The Main Query Form
in SPARQL
• It has two basic components: SELECT ?band
WHERE {
1. A list of selected variables
?band rdf:type :Band .
2. Triple patterns to match }

• Results are returned as a table

SELECT ?album ?artist
where each selected variable is WHERE {
?album rdf:type :Album .
a column and each pattern
?album :artist ?artist .
match is a row ?artist rdf:type :SoloArtist .
}
Single Triple Pattern

SELECT ?album
WHERE {
?album rdf:type :Album .
}

SELECT *
{
?album a :Album
}
Joins

SELECT * {
?album a :Album .
?album :artist ?artist .
}

SELECT * {
?album a :Album .
?album :artist ?artist .
?artist a :SoloArtist .
}
Optional Join

SELECT ?song ?length {

?song a :Song .
OPTIONAL {
?song :length ?length .
}
}
Subqueries

SELECT (avg(?count) AS ?avgCount)

{
SELECT ?year (count(?album) AS ?count)
{
?album a :Album ;
:date ?date .
BIND (year(?date) AS ?year)
}
GROUP BY ?year
}
Alternatives

SELECT ?name
{
{ ?artist a :SoloArtist }
UNION
{ ?artist a :Band }
?artist :name ?name
}
Negation

SELECT ?song {
?song a :Song .
FILTER (
NOT EXISTS {
?song :length ?length .
}
)
}
Sort Results

SELECT *
{
?album a :Album ;
:artist ?artist ;
:date ?date
}
ORDER BY ?date
Limit Results

SELECT *
{
?album a :Album ;
:artist ?artist ;
:date ?date
}
ORDER BY ?date
LIMIT 2
Offset Results

SELECT *
{
?album a :Album ;
:artist ?artist ;
:date ?date
}
ORDER BY ?date
LIMIT 2
OFFSET 2
Filtering Results

SELECT *
{
?album a :Album ;
:artist ?artist ;
:date ?date
FILTER (year(?date) >= 1970)
}
ORDER BY ?date
Binding Variables

SELECT *
{
?album a :Album ;
:artist ?artist ;
:date ?date
BIND (year(?date) AS ?year)
FILTER (?year >= 1970)
}
ORDER BY ?date
Removing Duplicates

SELECT DISTINCT ?year

{
?album a :Album ;
:artist ?artist ;
:date ?date
BIND (year(?date) AS ?year)
}
ORDER BY ?year
Grouping Results

SELECT ?year (count(?album) AS ?count)

{
?album a :Album ;
:date ?date ;
BIND (year(?date) AS ?year)
}
GROUP BY ?year
ORDER BY desc(?count)
Query Types
ASK Query

ASK {
?band a :Band .
?song :writer ?band .
}
CONSTRUCT Query

CONSTRUCT WHERE {
?band a :Band ;
:member ?member
}

CONSTRUCT {
?member a :BandMember
}
WHERE {
?band a :Band ;
:member ?member
}
Named Graphs
RDF Datasets

• An RDF dataset is a collection of RDF graphs: Default Graph (no IRI)

• There is exactly one default graph

Named Graph 1
<...#g1>
• It does not have a name
• May be empty or contain RDF triples Named Graph 2
<...#g2>
• Zero or more named graphs
Named Graph 3
• A named graph is an RDF graph <...#g3>

identified by an IRI
Named Graph 4
• Graph names are unique within an <...#g4>

RDF dataset
RDF Data in TriG: Turtle with Named Graphs
• RDF data for the default
GRAPH :Artist {
graph and zero or more :The_Beatles a :Band .

named graphs can be ...

}
serialized in a single
GRAPH :Album {
document :Please_Please_Me rdf:type :Album .
• Use GRAPH to specify a ...

named graph followed by its }

}
triples
 Specifying SPARQL Dataset
• A query can use FROM to override the default
PREFIX ex: <...>
SELECT * graph and temporarily treat the merge of one
FROM ex:g1
FROM ex:g4 or more graphs as though they are the
FROM NAMED ex:g1 default graph
FROM NAMED ex:g2
FROM NAMED ex:g3 • FROM NAMED determines which graphs can
WHERE {
...Pattern A... be available as named graphs in the query
GRAPH ex:g3 { • GRAPH directs a query either to a particular
...Pattern B...
} named graph, or to any of the available
GRAPH ?graph {
named graphs
...Pattern C...
}
} Based on: http://www.slideshare.net/LeeFeigenbaum/sparql-cheat-sheet
 Specifying SPARQL Dataset
In this example…
PREFIX ex: <...>
SELECT * • Pattern A results come from the merge of g1
FROM ex:g1
FROM ex:g4 and g4 which temporarily act as the default
FROM NAMED ex:g1 graph
FROM NAMED ex:g2
FROM NAMED ex:g3 • Pattern B results can only come from the
WHERE {
...Pattern A... named graph g3
GRAPH ex:g3 { • Pattern C results may come from any of the
...Pattern B...
} named graphs available to this query: g1, g2, or
GRAPH ?graph {
g3. The ?graph variable will specify the
...Pattern C...
} source(s) of any results
} Based on: http://www.slideshare.net/LeeFeigenbaum/sparql-cheat-sheet
 Querying a Specific Dataset

SELECT * {
default graph Default Graph (no IRI)
...graph pattern... #
and any named
graphs
GRAPH :g1 {
...graph pattern... # a specific graph Named Graph 1
<...#g1>
}
GRAPH ?graph { Named Graph 2
or <...#g2>
...graph pattern... # any named graph
or
} Named Graph 3
<...#g3>
}
Based on: http://www.slideshare.net/LeeFeigenbaum/sparql-cheat-sheet
Named Graph Query

SELECT * {
GRAPH :Album {
?album a :Album .
?album :artist ?artist .
}
GRAPH :Artist {
?artist a :SoloArtist .
}
}