Rule-based reasoning in the Semantic Web

Rule-based reasoning in
the Semantic Web

Fulvio Corno, Laura Farinetti
Politecnico di Torino
Dipartimento di Automatica e Informatica
e-Lite Research Group – http://elite.polito.it

Outline
 Motivation
 SWRL
 RIF

F. Corno, L. Farinetti - Politecnico di Torino 2

Motivation
 Classical reasoning operations in OWL only allow
very limited use cases
 Consistency
 Class properties and relationships
 Instance classification
 How to to more complex reasoning tasks?
 E.g., inferring properties of instances depending on
other instance properties?


Types of rules
 Production Rules (or Inference Rules)
 IF condition THEN action
 Logical predicates
 Executed upon explicit request, starting from a
stateless initial condition
 Event-Condition-Action rules
 “Reactive” execution to detect and react to incoming
events and process event patterns
 Usually running in parallel to the application, reacting
when events occur, usually in a stateful manner


Rule-based reasoning
 Forward chaining
 Backward chaining
 RETE Algorithm
 Goal-directed reasoning


Forward chaining
 Forward chaining starts with the available data and
uses inference rules to extract more data until
a goal is reached.
 An inference engine using forward chaining searches
the inference rules until it finds one where
the antecedent (If clause) is known to be true.
 When found it can conclude, or infer,
the consequent (Then clause), resulting in the
addition of new information to its data


Backward chaining
 Backward chaining starts with a list of goals (or
a hypothesis) and works backwards from
the consequent to the antecedent to see if there
is data available that will support any of these
consequents.
 An inference engine using backward chaining would
search the inference rules until it finds one which has
a consequent (Then clause) that matches a desired
goal.
 If the antecedent (If clause) of that rule is not known
to be true, then it is added to the list of goals (in order
for one's goal to be confirmed one must also provide
data that confirms this new rule).

Rule engines
 Drools (now Jboss Rules, own rule format)
 http://www.jboss.org/drools
 SweetRules (RuleML, SWRL, others)
 http://sweetrules.semwebcentral.org/
 Jena 2 (own rule format)
 http://jena.sourceforge.net/inference/
 Jess (own rule format)
 http://www.jessrules.com/
 Pellet
 Bossam
…

Outline
 Motivation
 SWRL
 RIF


Semantic Web Rule Language
(SWRL)
 Not an official W3C Recommedation
 Application to OWL of the RuleML (http://ruleml.org/)
languages
 Extends OWL language by providing Horn clauses
 Defines an extension of the OWL model-theoretic
semantics


SWRL structure
 The rules are of the form of an implication between an
antecedent (body) and consequent (head).
 The intended meaning can be read as:
 whenever the conditions specified in the antecedent
hold,
 then the conditions specified in the consequent must
also hold.


General structure
 Both the antecedent (body) and consequent (head)
consist of zero or more atoms.
 An empty antecedent is treated as trivially true (i.e.
satisfied by every interpretation), so the consequent
must also be satisfied by every interpretation;
 an empty consequent is treated as trivially false (i.e.,
not satisfied by any interpretation), so the antecedent
must also not be satisfied by any interpretation.
 Multiple atoms are treated as a conjunction


Atoms
 Atoms in these rules can be of the form
 C(x), where C is an OWL description
 P(x,y), where P is an OWL property
 sameAs(x,y)
 differentFrom(x,y)
 where x, y are either variables, OWL individuals or
OWL data values


Syntax issues
 SWRL rules are defined according to different syntax
forms
 Abstract
syntax (in functional form)
 XML concrete syntax
 RDF concrete syntax
 Human-readable form (using logic predicates)


Example: uncle
 Human-readable syntax
 hasParent(?x1,?x2) ∧ hasBrother(?x2,?x3) ⇒
hasUncle(?x1,?x3)
 Abstract syntax
 Implies(Antecedent(
hasParent(I-variable(x1) I-variable(x2))
hasBrother(I-variable(x2) I-variable(x3)))
Consequent(
hasUncle(I-variable(x1) I-variable(x3))))
 Example: if John has Mary as a parent and Mary has
Bill as a brother then John has Bill as an uncle


Example: inheritance
 Student(?x1) ⇒ Person(?x1)
 Abstract syntax
 Implies(Antecedent(Student(I-variable(x1)))
Consequent(Person(I-variable(x1))))

 This is an improper usage of rules: it should be
expressed directly in OWL, to make the information
also available to an OWL reasoner
 SubClassOf(Student Person)


Example: propagating properties
 Artist(?x) & artistStyle(?x,?y)
& Style(?y) & creator(?z,?x) ⇒
style/period(?z,?y)
 Abstract syntax
 Implies(Antecedent(
Artist(I-variable(x))
artistStyle(I-variable(x) I-variable(y))
Style(I-variable(y))
creator(I-variable(z) I-variable(x)))
Consequent(style/period(I-variable(z) I-
variable(y))))
 Meaning: the style of an art object is the same as the
style of the creator

SWRL versus OWL
 The last example may not be described in OWL
 In OWL, you declare relationships between Classes
 Such relationships are intended to apply on instances
 You may add properties to instances to materialize
such relationships
 OWL Inference only supports “forall” or “exists” in
propagating properties
 In OWL you may not express “that specific instance
that has such properties”!


OWL versus SWRL
 OWL has a declarative nature, while SWRL is more
operational
 Even if the semantics extends that of OWL, practical
reasoners just “apply the rules”
 The consistency of the rules application relies on the
rule designer’s infinite wisdom
 Example: If a property is declared as symmetric, then
we must be careful to create all property instances to
satisfy that


Concrete syntaxes: XML
<ruleml:imp>
<ruleml:_rlab ruleml:href="#example1"/>
<ruleml:_body>
<swrlx:individualPropertyAtom swrlx:property="hasParent">
<ruleml:var>x1</ruleml:var>
</swrlx:individualPropertyAtom>
<swrlx:individualPropertyAtom swrlx:property="hasBrother">
</ruleml:_body>
<ruleml:_head>
<swrlx:individualPropertyAtom swrlx:property="hasUncle">
</ruleml:_head>
</ruleml:imp> Politecnico di Torino
F. Corno, L. Farinetti - 20

Concrete syntaxes: RDF (I)
<swrl:Variable rdf:ID="x1"/>
<ruleml:Imp>

<ruleml:body rdf:parseType="Collection">

<swrl:IndividualPropertyAtom>
<swrl:propertyPredicate rdf:resource="&eg;hasParent"/>
<swrl:argument1 rdf:resource="#x1" />
</swrl:IndividualPropertyAtom>

<swrl:propertyPredicate rdf:resource="&eg;hasSibling"/>

Concrete syntaxes: RDF (II)
<swrl:propertyPredicate rdf:resource="&eg;hasSex"/>
<swrl:argument2 rdf:resource="#male" />

</ruleml:body>

<ruleml:head rdf:parseType="Collection">

<swrl:propertyPredicate rdf:resource="&eg;hasUncle"/>

</ruleml:head>
</ruleml:Imp>

Outline
 Motivation
 SWRL
 RIF


Source: Chris Welty (W3C), RIF
Presentation to NY Semweb Meetup,
F. Corno, L. Farinetti - Politecnico di Torino
http://www.w3.org/2005/rules/wiki/images
24
/b/b0/W3C_RIF-CW-9-09.pdf

References
 SWRL: A Semantic Web Rule Language
Combining OWL and RuleML, W3C Member
Submission 21 May 2004
 http://www.w3.org/Submission/SWRL/
 RIF Use Cases and Requirements, W3C Working
Draft 18 December 2008
 http://www.w3.org/TR/rif-ucr/


References
 http://en.wikipedia.org/wiki/Business_rules_engine
 http://en.wikipedia.org/wiki/Forward_chaining
 http://en.wikipedia.org/wiki/Backward_chaining
 RIF Working Group
 http://www.w3.org/2005/rules/wiki/RIF_Working_Group


License
 This work is licensed under the Creative Commons
Attribution-Noncommercial-Share Alike 3.0
Unported License.
 To view a copy of this license, visit
http://creativecommons.org/licenses/by-nc-sa/3.0/ or
send a letter to Creative Commons, 171 Second
Street, Suite 300, San Francisco, California, 94105,
USA.


Rule-based reasoning in the Semantic Web

Recommandé

Recommandé

Contenu connexe

Tendances

Tendances (8)

Similaire à Rule-based reasoning in the Semantic Web

Similaire à Rule-based reasoning in the Semantic Web (20)

Dernier

Dernier (20)

Rule-based reasoning in the Semantic Web