To jest stara wersja strony!


Opis

Weronika Furmańska ikaf@student.agh.edu.pl

Porównać i omówić metody reprezentowania atrybutów w językach:

oraz metody reprezentowania reguł w:

  • RIF
  • R2ML
  • RuleML

co to jest datalog

business rules orchestration

tools: http://sweetrules.projects.semwebcentral.org/

people http://www.mit.edu/~bgrosof/

concepts: http://en.wikipedia.org/wiki/Description_Logics

http://www.w3.org/2005/rules/wg/wiki/negation

http://www.w3.org/2005/rules/wiki/RIF_Working_Group

Spotkania

Projekt

Draft…

Part 1. Attributes representation in different markup languages

To start with...

While talking about attribute-based languages we can recognize the following types of logic:

  • AAL – Attributive Logic, i.e. attributive logic with atomic values of attributes only
  • SAL – Set Attributive Logic, i.e. attributive logic with set values of attributes
  • VAAL – Variable Atomic Attributive Logic, i.e.attributive logic with atomic values of attributes incorporating variables
  • VSAL - Variable Set Attributive Logic, i.e.attributive logic with set values of attributes incorporating variables

The alphabet and the crucial question:What is an attribute?

Let there be the following, pairwise disjoint :-) sets of symbols:

  • O – a set of object name symbols
  • A – a set of attribute names
  • D – a set of attribute values names (the domains)
  • V – a set of variable symbols (variables)

If the set of attributes is finite and given as
:pl:miw:atrybuty.png
then
:pl:miw:domain.png,
where Di is the domain of attribute Ai , i=1, 2, …,n.

Attributes in AAL and VAAL
An attribute Ai is a function (or partial function) of the form:
:pl:miw:atrybut-def.png
An attribute can be a partial function, since not all the attributes must be defined for all the objects.

Generalized attributes in SAL and VSAL
A generalized attribute Ai is a function (or partial function) of the form
:pl:miw:atrybut-def2.png

RDF - Resource Description Framework

Basic information

  1. RDF identifies things using Web identifiers (URIs)
  2. RDF is written in XML. The language is called RDF/XML.
  3. The main elements of RDF are the root element, <RDF>, and the <Description> element, which identifies a resource.
  4. <rdf:RDF> is the root element of an RDF document. It defines the XML document to be an RDF document. It also contains a reference to the RDF namespace. <rdf:RDF xmlns:rdf=„http://www.w3.org/1999/02/22-rdf-syntax-ns#
  5. The <rdf:Description> element identifies a resource with the rdf:about attribute. This element contains elements that describe the resource.

Attributes representation

„RDF is based on the idea that the things being described have properties which have values, and that resources can be described by making statements that specify those properties and values. RDF uses a particular terminology for talking about the various parts of statements.” (W3C RDF Primer) Attributes are represented in RDF in a form of RDF Triples (subject, predicate, object) where subject denote the resource being described, predicate indicate the name of the attribute and object points to the attribute (predicate) value. RDF statements (triples) consist of subject, predicate and object which all are identified by URIs.

Part of an RDF triple Can be
Subject URI, blank node
Predicate URI
Object URI, literal (constant values, plain or typed), blank node


RDF uses VSAL

http://www.w3.org/TR/rdf-concepts/#dfn-blank-node

  • sets of attributes cannot be defined in a single RDF triple, because attributes in RDF are represented as binary predicates only. However, it is possible to do that using RDF Container Elements (<rdf:Bag> and <rdf:Seq>) or RDF Collections. To define sets of attributes we create a new resource (which may either be a blank node or a resource with a URIref). The resource is given an rdf:type property whose value is one of the predefined resources rdf:Bag, rdf:Seq, or rdf:Alt . Then the members of the container can be described by defining a container membership property for each member with the container resource as its subject and the member as its object.

http://www.w3.org/TR/rdf-primer/#containers

RDF/XML notation

Triples of the Data Model (an example from W3Schools.):

To avoid using URI all around we use namespaces which we define as follows: xmlns:si=„http://www.recshop.fake/siteinfo#
In RDF/XML attributes (properties of subjects) can be defined in three ways:

  • Properties as XML elements
<?xml version="1.0"?>
 
<rdf:RDF
xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" <!-- rdf namespace -->
xmlns:si="http://www.recshop.fake/siteinfo#"><!-- other namespace (eg. Our own website's)>
  <rdf:Description rdf:about="http://www.w3schools.com/RDF"> <!-- subject identifed with a resource by rdf:about attribute -->
    <si:author>Jan Egil Refsnes</si:author> <!-- property and its value -->
    <si:homepage>http://www.w3schools.com</si:homepage> <!-- another property and its value -->
  </rdf:Description>
</rdf:RDF>

When you parse the example above with a RDF Validator, the result will look something like this

  • Properties as attributes
  • Properties as resources

Do powyższych podpunktów nie mogłam dołączyć kodu - otrzymuje błąd, że metoda POST jest niezaimplementowana :-/ prawdopodobnie sekcja tekstu jest za długa. — Weronika Furmańska 2008/05/21 22:35

ERDF - Extended RDF

Basic information

ERDF extends RDF graphs with weak and strong negation, as well as derivation rules. The main concepts of the ERDF are based on Partial Logic.

Attribute representation

Partial Logic allows to distinguish between properties and classes that are completely represented in a knowledge base and those that are not. It is to the owner of the database to classify whether for a given property there is a complete information or not. On completely represented properties (which are by design treated as partial ones) the close-world assumption can be applied and the entailment of :pl:miw:weak_negation.png allows to derive :pl:miw:strong_negation.png. However, we can declare a property to be total. If we do so, the property can be both true or false under some interpretation and the rule „:pl:miw:weak_negation.png derives :pl:miw:strong_negation.png” is not applicable. There has to be an explicit negative information about a total property to infer negated statements about it.

ERDF let us define if a property or a class as total (or partial by default) and therefore support both close-world and open-world reasoning.

ERDF extends RDF and RDFSchema in a couple of ways.

  1. First, RDF triples are extended with the inclusion of strong negation, so one can specify a predicate which is explicitly negative.
    <ex:likes erdf:negationMode="Sneg" rdf:resource="#Pork"/> 
  2. RDFS vocabulary (which let one define classes and properties) are extended so that we can declare a property or class to be total (and thus enable open-world reasoning on them).
  3. There can be a plain literal in the place of ERDF triple subject. So now it looks as follows:
Part of an ERDF triple can be
Subject URI or literal (plain or typed) or a variable
Predicate URI
Object URI or literal (plain or typed) or a variable

ERDF/XML notation

ERDF/XML syntax folllows the RDF/XML one and extends it in a suitable way.

RDFS is extended with the terms erdf:TotalClass and erdf:TotalProperty. We can define a property p or a class c to be total using

rdf:type(p,erdf:TotalProperty)

and

rdf:type(c,erdf:TotalClass)

statements.

Example of a regular (partial by design) property:

<rdf:Property rdf:about="#likes">
  <rdfs:domain rdf:resource="#Person"/>
</rdf:Property>

Example of a total property:

<erdf:TotalProperty rdf:about="#authorOf">
  <rdfs:domain rdf:resource="#Person/">
  <rdfs:range rdf:resource="#Book"/>
</erdf:TotalProperty>

ERDF triples (and sets of ERDF triples sharing the same subject term) are encoded by means of the erdf:Description element, which has a non-empty list of (possibly negated) property-value slots about the subject term. The following rules apply:

  • URI references, blank node identifiers and variable that appear in the subject position are expressed as values of the erdf:about attribute:
    <erdf:Description erdf:about="#Gerd">
    ...
    </erdf:Description> 
  • literals that appear in the subject position are expressed as the text content of the erdf:about subelement:
    <erdf:Description>
      <erdf:About rdf:datatype="&xsd;string">Grigoris</erdf:About>
      ...
    </erdf:Description> 
  • URI references, blank node identifiers and variable that appear in the object position are expressed as values of the attributes rdf:resource, rdf:nodeID and rdf:variable respectively:
     <ex:denotationOf f:resource="#Grigoris"/> 
  • literals that appear in the subject position are expressed as the text content of the corresponding subelement

Sprawozdanie

Materiały

pl/miw/piw08_hekateonto.1212075405.txt.gz · ostatnio zmienione: 2019/06/27 15:59 (edycja zewnętrzna)
www.chimeric.de Valid CSS Driven by DokuWiki do yourself a favour and use a real browser - get firefox!! Recent changes RSS feed Valid XHTML 1.0