ONTOLOGY-BASED INTEGRATION OF XML DATA
Schematic Marks as a Bridge Between Syntax and Semantic Level
Christophe Cruz, Christophe Nicolle
Laboratoire LE2I, UMR CNRS 5158,Université de Bourgogne, BP 47870, 21078 Dijon Cedex – France
Keywords: XML, XML schema, Ontology, Integration, Semantic.
Abstract: This paper presents an ontology integration approach of XML data. The approach is composed of two
pillars the first of which is based on formal language and XML grammars analysis. The second pillar is
based on ontology and domain ontology analysis. The keystone of this architecture which creates a bridge
between the two pillars is based on the concept of schematic marks introduced in this paper. These
schematic marks make it possible to establish the link between the syntactic level and the semantic level for
our integration framework.
1 INTRODUCTION
Data integration consists in inserting a data set into
another data set. In the context of the Web,
document integration consists in creating
hypermedia links between the documents using
URL. Associated documents are multimedia
documents which can be text, pictures, videos,
sounds or any other file format. In the context of
database integration, integrated data coming from
several information systems makes information
more complete and more relevant with a more global
objective use. For example the biomedical data
source is known to be hyper-linked because the
description of objects suggests several hyperlinks,
allowing the user to “sail” from one object to
another in multiple data sources. Indeed there are on
the Web more than one hundred genetic databases,
two hundred twenty-six data sources of molecular
biology, etc. (A. Baxevanis, 2000),(D. Benson et al.,
2000). A second example of activity which is more
and more based on the use of complex data source
integration is the field of geographical information
management. Just as biological and medical data,
geographical data are heterogeneous and distributed.
These data are for example weather (France:
www.meteofrance.com) or cartographic (road maps:
www.viamichelin.com). In fact data integration of
various sources will thus make it possible to carry
out more complex, precise requests and improve the
various information systems available. The objective
of data integration is to benefit from the diversity of
information available and to benefit from the rise of
new Web technologies. Moreover the re-use of data
and services make it possible to optimize the costs
for the acquisition and the maintenance of
information. Finally, the management and the cost of
the data-processing resources are distributed among
the whole of the data suppliers.
2 BACKGROUND
In the context of the Web, XML technologies
became a headlight technology for data structuring
and data exchanges. Many systems using XML as
databases integration have a mediation approach (A.
Pan et al.,2002), (D. Draper et al., 2001), (M. J.
Carey et al., 2000), (A. Cali et al., 2001). The
evolution of the Web technologies changed the
integration problem of information. In fact the XML
contribution to define not only integration schemas
but also the definition languages of the
corresponding models reduced considerably the
problems related to the structural and the syntactic
heterogeneity. The contribution of the Web
technologies related to the service oriented
architectures solved partially the problems of the
localization and the data access allowing the design
of interoperability architectures on a greater scale
(K. Aberer et al., 2001). Nevertheless, during the
integration data process and the integration services
there remain many problems related to semantic
heterogeneity.
In order to support the action of agents,
knowledge has to represent the real world by
reflecting entities and relations between them.
30
Cruz C. and Nicolle C. (2006).
ONTOLOGY-BASED INTEGRATION OF XML DATA - Schematic Marks as a Bridge Between Syntax and Semantic Level.
In Proceedings of WEBIST 2006 - Second International Conference on Web Information Systems and Technologies - Internet Technology / Web
Interface and Applications, pages 30-37
DOI: 10.5220/0001245400300037
Copyright
c
SciTePress
Therefore knowledge constitutes a model of the
world and agents use their knowledge as a model of
the world. On the one hand the integration of
different entities is possible when the semantic of
the entities and the relations is identical. In addition,
to model the semantic of knowledge as well as the
structure where this knowledge is stored, it is
necessary to reach a higher conceptual level. For that
knowledge representation is independent of
knowledge use. Thus knowledge representation and
inferential mechanisms are dissociated (N. Guarino
et al., 1994). On the other hand, domain
conceptualization can be performed without
ambiguity only if a context of use can be given. In
fact a word or a term can designate two different
concepts depending on the particular context of use
(B. Bachimont et al., 2000). The semantic of
knowledge is strongly constrained by the symbolic
representation of computers. Therefore N. Guarino
(N. Guarino, 1994) introduced an ontological level
between the conceptual level and epistemological
level. The ontological level forms a bridge between
interpretative semantics in which users interpret
terms and operational semantics in which computers
handle symbols (T. Dechilly and B. Bachimont,
2000). In fact the problem consists in defining an
upper conceptual level to handle the semantic in
XML documents for their integration. This level will
define a semantic framework leading to the
integration of XML data by the use of an ontology.
The implementation of an ontology is a mapping
stage between the system elements and their
ontological "counterparts". Once this mapping has
been carried out, the representation of elements in
the ontology is regarded as a meta-data diagram. The
role of a meta-data diagram is double (B. Amann
and D. Partage, 2003). On the one hand it represents
an ontology of the knowledge shared on a domain.
On the other hand it plays the role of a database
schema which is used for the formulation of requests
structured on meta-data or to constitute views. This
principle is applied to ontology based data
integration using domain ontology to provide
integration structures and request processes to these
structures. According to (I. F. Cruz et al., 2004), (M.
Klein, 2002), (L. V. Lakshmannan and F. Sadri,
2003) data integration consists in defining rules of
mapping between information sources and the
ontological level. The principle consists in labeling
source elements and thus providing semantic
definition to elements compared to a consensual
definition of the meaning. This phase is inevitably
necessary because this information was not added to
the document during its creation. Moreover an XML
schema defines only the structure of associated
XML documents. However, an XML schema
contains tacit knowledge that can be used to define
ontology by extracting a set of elements and
properties whose meaning will be defined for a more
global use.
Section 3 describes a general view of our method
based on two pillars (ontology and formal language).
The keystone of our method is the concept of
schematic marks. This section describes this concept
by a formal way. Section 4 defines a set of
integration rules based on the schematic marks.
3 METHOD OVERVIEW
Our integration solution consists in connecting
various levels of semantic and schematic abstraction.
This solution is articulated in two stages. The first
stage relates to the semantic formalization of the
writing rules to define an XML grammar. This
formalization will enable us to define the
components of a generic ontology. The second stage
relates to the definition of the ontologization
mechanisms of the semantic elements from a
specific XML grammar to obtain an ontology of
domain. The concepts and the relations of the
domain ontology are then defined starting from the
elements of the XML schema. These mechanisms
make it possible to identify some concepts and
relations common to several XML schemas.
Consequently ontology makes it possible to link the
concepts and the relations by amalgamating the
attributes of the common elements which are
semantically identical. The domain ontology will be
extended and then modified to represent the
semantic of several XML schemas relating to a
particular domain. To specify the “semantic
elements” of an XML schema it is first necessary to
identify and mark them. We call these schematic
marks. They will be used to establish links between
the structure of an XML document and its semantic
definition. Those schematic marks represent the
structural level of the integration system.
Figure 1: The two pillars and the keystone of our method.
To specify the semantic of the XML schema
elements it is necessary to identify and mark them
using schematic marks. These marks will be used to
establish links between the structure of the XML
document and its semantic definition. This section
presents first of all the formalization of XML
ONTOLOGY-BASED INTEGRATION OF XML DATA - Schematic Marks as a Bridge Between Syntax and Semantic
Level
31
documents using formal languages. In addition this
section outlines a formalization of XML grammars
using the formal languages on which the principle of
schematic marks is based. This underlines the fact
that XML grammars generate languages of Dyck.
According to definite properties of Dyck languages
the concepts of factor and schematic mark are
defined.
3.1 XML Document Formalization
An XML document is composed of text and opening
tags associated to closing tags. Some of these tags
are at the same time opening and closing tags. In
fact, empty tags define the sheets of the XML trees.
One of the properties of an XML canonical
document is to be composed of only opening or
closing tags. Empty tags can be exchanged by
opening and closing tags without any problems for
the XML parser. Consequently, any XML file
having empty tags has an equivalent without empty
tags. This property is syntactic because it does not
appear in the grammatical rules formalizing the
structure of the document. Starting from this
information some definitions are expressed.
Definition 1: An alphabet is a finite set of
symbols recorded
Σ . These elements are called
letters. In this paper most of the time it will be
written:
{}
…,,ba=Σ . The size Σ of an alphabet
Σ equals the number of its elements.
Definition 2: A word or a sentence on
Σ
is a
sequence of letters coming from this alphabet. The
word « wall » is a sequence of letters from the
alphabet
{}
zba ,,, …=Σ . It is said by convention
that the word empty is the null size word. It is
written:
ε
. The set of all words that are possible to
be written on the alphabet
Σ is written:
*
Σ
.
{}
ε
−Σ=Σ
+ *
.
Definition 3: A formal language
L
is a set of
words on
*
Σ . A language L on the alphabet
Σ
is
called regular (A regular language is a language of
kind 3 in the Chomsky hierarchy) only if it is
generated on the alphabet
Σ
and if it is defined by a
regular expression. It means that the set of regular
languages on
Σ is exactly the set of languages
recognizable by finite state automaton on
Σ
. In
others words for each finite state automaton it is
possible to associate regular expressions that define
identical languages recognized by the same
automaton and reciprocally.
We have just seen that a language L is made of
words generated starting from an alphabet. If in an
XML document we just consider the tree structure
without taking into account the values of the tags’
attributes, then the set of the XML documents which
it is generable from a XML schema define a
language. Moreover the set of the tags of XML
documents defined by XML schema represents a
part of the alphabet on the language. It defines only
one part because the alphabet can contain letters not
used by the language.
Definition 4: A formal grammar can be defined
as a mathematical entity on which we can associate
an algorithmic process to generate a language. It is
defined as a quadruplet
SPTNG ,,,=
in which:
•
T
written also
Σ
is the terminal alphabet
on
G .
•
N the non terminal alphabet onG .
•
TNV ∪
=
is an alphabet composing the
whole symbols of
G .
•
P
is a set of production rules or regular
expressions.
•
NS
∈
is the start symbol on
G
.
{
}
ASN ,=
{
}
baT ,=
(
)
(
)
(
)( )
{
}
bbAaaASAASP →→→→
=
,,,
ε
bbaaA
AAS
→
→
ε
SPNG ,,,
1
Σ=
Example 1: G
1
grammar.
Example 1 shows a grammar for which the
generated words are «aaaa» «aabb» «bbaa» «bbbb»
and «». The language is composed of four words and
the empty word.
3.2 Formal Grammar and XML
Grammar
We saw in the preceding section the concept of
regular language and that of formal grammar. This
section presents formal grammars and XML
grammars by making connections between them.
These grammars have the characteristic to have a
WEBIST 2006 - INTERNET TECHNOLOGY
32
final vocabulary composed of opening tags and
closing tags.
Definition 1: For a given set
A
composed of
opening tags and corresponding closing tags, an
XML document is a word composed from the
alphabet
AA
T
∪= .
For the moment we just take into account the
syntactic structure. An XML document x is well
formed if only one tag is root and if the tags are
correctly imbricated.
Definition 2: A document x is well formed if x is
generated by production rules from a language of
Dyck on
A
A
T
∪=
. A language of Dyck is a
language generated by a context-free grammar
where
Aa
n
∈
and
Ab
n
∈
:
1
n2211
≥→ nwithSbaSbaSbaSSS
n
…
ε
This definition corresponds to the terminology
and the notation used in the XML community. A
language is indeed a set of XML documents which
can be generated starting from a grammar. These
grammars of XML languages are called "Document
Type Definition" (DTD). The axiom of grammars is
qualified DOCTYPE and the whole of the
production rules is associated to a tag ELEMENT
(Example 2). A tag ELEMENT is made up of a type
and a model. The type is simply the name of the tag
and the model is the regular expression for the
element.
()()()
(
)
(
)
{
}
bbTbbTTaTSTSaSP →→→= ,,
<?xml version="1.0" encoding="UTF-8"?>
<!ELEMENT a ((a | b), (a | b))>
<!ELEMENT b (b*)>
Example 2: Similarity between a grammar and a DTD.
All of the production rules corresponding to the
grammar can also be represented using a XML
schema which can be translated out of a DTD
(Example 3). DTD only defines the relations
between the various components of a document
contrary to the XML schema that defines also data
types.
<?xml version="1.0" encoding="UTF-8"?>
<!ELEMENT ROOT (A | (B, C))>
<?xml version="1.0" encoding="UTF-8"?>
<xsd:schema
xmlns:xsd="http://www.w3.org/2001/XMLSchema"
elementFormDefault="qualified">
<xsd:element name="A"/>
<xsd:element name="B"/>
<xsd:element name="C"/>
<xsd:element name="ROOT">
<xsd:complexType>
<xsd:choice>
<xsd:element ref="A"/>
<xsd:sequence>
<xsd:element ref="B"/>
<xsd:element ref="C"/>
</xsd:sequence>
</xsd:choice>
</xsd:complexType>
</xsd:element>
</xsd:schema>
Example 3: Similarity between DTD and W3C XML
schema.
In this section we saw reminders on the formal
languages by making parallels between a formal
grammar and an XML schema. We know that an
XML schema is a formal grammar that generates a
language of Dyck and that it has consequently the
properties of a language of Dyck. The following
section describes the properties of grammars that
generate languages of Dyck and introduces the
definition of schematic marks.
3.3 Factor and Schematic Marks
The properties of the languages of Dyck were the
subject of studies undertaken by J Berstel (J. Berstel
and L. Boasson, 2000). By drawing parallels
between XML grammar and the languages of Dyck,
J Berstel defines the concept of factor. According to
the lemma 3.3 of J.Berstel if
G is an XML
grammar on
A
A
T ∪
=
generating a language
L
with a non terminal vocabulary
a
X
and
Aa ∈
then
for each
Aa
∈
the language generated by
a
X
is a
set of factors of words in
L
which are languages of
Dyck starting by the letter
a
:
() ()
LFXL
aaG
=
A given language:
(
)
(
) ()
{
}
0n
1,2
21
>=
k
nnn
cabbaabbaabbcaL
k
……
then
() ()
{
}
()
()
{
}
abbaLFbbLFLLF
abc
**
, , ===
Example 4: Factors of the language L.
ONTOLOGY-BASED INTEGRATION OF XML DATA - Schematic Marks as a Bridge Between Syntax and Semantic
Level
33
This means that a language is factorizable in an
under language and a factor of a language of Dyck is
a language of a language of Dyck. Consequently an
under tree of an XML document can be generated by
a factor of the language of Dyck to which the XML
document belongs.
According to the corollary 3.4 of J. Berstel there
is only one factor for an XML grammar
(
)
LLF
a
=
.
This means that there is only one ‘father’ tag for all
others. This tag is the root. Consequently there is
only one factor
()
LLF
a
= where
a
is the root.
According to the same corollary 3.4: For a given
word
w of the language L there is a unique
factorization
auuauw
n
aaa
…
21
= with
ii
aa
Du ∈
for
ni …,1∈
.
i
a
D
is a language of
Dyck starting by
i
a and
Aa
DD
i
⊂
. The trace of a
word
w is defined by a word
n
aaa …
21
. The
surface
Aa ∈ in the language
L
is a set
(
)
LS
a
of all traces of the words of the factor
(
)
LF
a
. The
notion of surface is used by Berstel to demonstrate
the following proposition:
Proposition 1: For each XML language
L there
is only one reduced XML grammar generating
L
.
This means that all the languages of an XML
grammar are generated by the same reduced XML
grammar. This independently of the values of tag’s
attributes in the documents. And if we only take into
account the syntactic structure of XML documents.
This proposal implies that if a factor were defined on
an XML language then this factor would correspond
to a production rule of the reduced grammar XML
generating this language. This proposal makes it
possible to introduce the concept of schematic mark.
A reduced grammar does not have any useless non
terminal vocabulary which is, in general, not the
case for grammars generating languages of Dick.
But an XML schema does not use unnecessary tags,
so an XML schema does not use unnecessary non
terminal vocabulary.
Definition 3: A schematic mark is a mark on an
XML schema to identify a production rule.
Figure 2: Definition of a schematic mark for the floor.
Those schematic marks allow to make links
between an XML schema element and a factor in a
corresponding XML language and, as a result, all
XML documents that can be generated by the XML
schema. In the following example the elements
which are marked define a concept. Consequently,
the schematic marks allow to provide links between
the concept and these instances. Thus, the schematic
marks can also mark a relational element and an
attribute element. Those concept elements, relational
elements and attribute elements are used to define
the domain ontology of the XML schema. The
schematic marks are used to link the component of
the ontology and the element of the XML schema.
At the ontology level the schematic marks are said to
be the semantic definition of the XML schema
elements that allow to identify the elements that
have a tacit semantic and are not defined formally.
Figure 3: Definition of schematic marks.
The following section shows that several
schematics marks of various XML schemas can have
the same semantics defined by a common ontology.
In this case, the element properties corresponding to
the schematic marks are integrated within the same
concept defined in ontology. It is called semantic
integration of XML schema.
4 INTEGRATION RULES
This section is outlined in three parts. The first part
presents the formalization of the XML grammar
construction rules. This formalization enables us to
release a set of terms which will be used to build our
generic ontology. This generic ontology allows to
define the elements of the domain of each XML
schema to be integrated. The second part presents
the definition of generic ontology. This generic
ontology is used as model to create domain
ontologies. This formal structure makes it possible
to index the tacit knowledge contained in the XML
schema. This knowledge is then explicit because it
raises any ambiguity on the interpretation of the
XML document terms generated starting from an
integrated XML schema.
4.1 Schematic Formalization
This section presents a set of definitions and rules to
formalize the construction of XML grammars and to
Formal Language
XML Grammar
XML Schema
Concept Tags
Relation Tags
Property Tags
XML Schema
<xs:element name=“Floor">
<xs:complexType>
<xs:sequence maxOccurs="unbounded">
<xs:element ref=“Slab" minOccurs="0"/>
<xs:element ref=“Wall" minOccurs="0"/>
</xs:sequence>
<xs:attribute name="Label" type="xs:string" />
</xs:complexType>
</xs:element>
Formal Language
XML Grammar
XML Schema
Concept Tags
Relation Tags
Property Tags
Ontology
Generic Ontology
Ontology of Domain
Concepts
Relations
Properties
WEBIST 2006 - INTERNET TECHNOLOGY
34
define the concepts, relations and attributes of our
generic ontology.
Definition 1: The Factor
()
LF
a
of a language
L is a conceptual factor if it defines a concept. For
example, the factor
()
LF
Building
is a concept
because it defines the concept Building.
Rule 1: If the conceptual factor
()
1
LF
a
of a
language
1
L and the conceptual factor
(
)
2
LF
b
of a
language
2
L
handle the same semantic of a common
concept then the intension of the concept is define
with the help of the conceptual factors
()
1
LF
a
of
1
L
and
()
2
LF
b
of
2
L .
Definition 2: The Factor
()
LF
a
of a language
L is a relational factor if it defines a relation. For
example, if a wall contains a door then the relational
factor
()
LF
Contain
is a relation because it defines a
relation between a wall and a door.
Rule 2: If the factor
()
1
LF
a
of a language
1
L
and the factor
()
2
LF
b
of a language
2
L handle
a
common semantic of a relation then the intension of
the relation is define with the help of the relational
factors
()
1
LF
a
of
()
2
LF
b
of
2
L .
Definition 3: The Factor
()
LF
a
of a language L
is an attribute factor if it defines one or several
proprieties of a concept or a relation. For example, if
a wall has a geometrical shape then the attribute
factor
()
LF
Geometry
is an attribute because it defines
the geometry of a wall.
Rule 3: If the factor
()
1
LF
a
of a language
1
L is
an attribute factor then it is integrated in the
intension of the concept or the relation. For example
if a wall has a thermical propriety the geometry
attribute factor
()
1
LF
a
is integrated in the intension
of the concept wall. The attribute factor can be
integrated in several intensions of concept and
relation.
The rules and definitions 1, 2 and 3 make it
possible to define the conceptual factors, the
relational factors, and the factor attributes of a
language. These factors correspond to schematic
marks carried out the elements of an XML grammar.
A schematic mark in an XML grammar corresponds
to a factor in the language. By defining these factors
as conceptual, relational or as attributes, we give
them a semantic. This semantic is already carried by
the elements but this marking allows to make it
explicit. Moreover, the definition of their attributes
constitutes the intension of the concept or the
relation and thus this improves their definition. By
amalgamating the attributes of the diagrammatic
marks of various XML schemas through the same
concept or the same relation we carry out an
integration of concepts or relations. This integration
constitutes the integration of an XML schema. The
following rules and definitions define the particular
cases.
Rule 4: If
(
)
LF
a
is a relational factor of a
language L then
(
)
LF
a
links a conceptual factor
‘father’ to a set of conceptual factor ‘sons’.
Rule 5: If the trace of a conceptual factor
(
)
LF
a
is composed of a conceptual factor
(
)
LF
b
then the link between the two conceptual factors
(
)
LF
a
and
(
)
LF
b
is a relational factor
()
LF
rab
. For
example if the conceptual factor wall has a
conceptual factor ‘son’ door then there is a relational
factor between the conceptual factor wall and the
conceptual factor door.
Rule 6: If the trace of a conceptual factor
(
)
LF
a
is composed of a conceptual factor set
(
)
LF
α
with A
∈
α
(alphabet AA
T
∪= ) and if
the semantic of the relation is the same then the link
between
(
)
LF
a
and the set
()
LF
α
is a relational
factor
(
)
LF
r
α
, having for conceptual factor
‘father’
(
)
LF
a
and for conceptual factor ‘son’ set
(
)
LF
α
. For example the conceptual factor floor
has a conceptual factor wall, a conceptual factor
column, a conceptual factor beam and a conceptual
factor slab. If the signification of the link between a
floor and the elements wall, column, beam and slab
is the same then the relational factor between the
element wall and the other elements are of the same
kind.
Definition 4: The intension of a concept is
composed of a set of schematic marks on XML
grammars. Those schematics marks are connected to
several conceptual factors of the language generated
by the grammars.
Definition 5: The extension of a concept is
composed of set of instances. In the present case
those instances are called semantic elements having
a trace which is a set of XML trees.
Definition 6: The intension of a relation is
composed of a set of schematic marks on the XML
grammars. Those schematic marks are connected to
a relational factor of the language generated by the
grammars.
Definition 7: The extension of a relation is
composed of a set of instances. In the present case
ONTOLOGY-BASED INTEGRATION OF XML DATA - Schematic Marks as a Bridge Between Syntax and Semantic
Level
35
those instances called relational elements have a
trace which is a set of XML trees.
Definition 8: A factor defines a concept or
relation or an attribute of an element from an XML
schema. Consequently, a mark is a conceptual factor
or a relational factor or an attribute factor.
Rule 7: If two instances of a concept represent
the same object or the same relation then they can be
identified as equal. For example, an object wall has a
set of thermical properties and another object wall
has a geometrical shape definition. Those two walls
are the same object if they have an identifier that
allows to identify them as the same object.
These definitions and rules give a set of vocabulary
that composes the taxonomy of our system. This
vocabulary is composed of the following words:
concept, relation, attribute, conceptual factor,
relational factor, attribute factor, semantic element,
and relational element. Each word defines a concept
of our generic ontology and is linked to a class.
• The class concept is defined by properties
represented by its intension which is composed
of a list of conceptual factors.
• The class relation is defined by properties
represented by its intension which is composed
of a list of relational factors.
• The semantic and relational elements are classes
allowing the instantiation of objects coming from
XML documents.
• The whole object of the class semantic element
linked to an instance of the class concepts
represents the extension of the instance of
concept.
• The whole object of the class relational element
linked to an instance of the class relation
represents the extension of the instance of
relation.
• The instances of the class conceptual factors and
the class relational factors are references to the
schematic marks on XML grammars. Those
marks are XML documents extracted from XML
grammars. The traces are XML documents
extracted from XML documents to integrate.
We saw until now the definitions and the rules
for the integration of XML schemas using a generic
ontology. These rules make it possible to share the
properties of various XML schemas. This level of
integration is called schema integration level
because it gathers in the heart of the same concept or
relation various properties defining the intension of a
concept or a relation. On this level of integration
follows a second level of integration. This level is
called data integration level. The first level defines
the concepts, the relations, and the attributes which
will be instanced on the second level of integration
using the schematic marks.
4.2 Generic Ontology
Previous sections have presented a partial
formalization of XML grammars which makes it
possible to build a generic ontology. This generic
ontology allows to define a set of domain ontologies
corresponding to various integrated XML schemas.
Figure 4: Def. of the generic concepts of our generic
ontology.
Once implemented this generic ontology makes
it possible to define the concepts, the relations, and
the properties of several domain ontologies.
Figure 5: Definition of two domain ontologies and
integration of the concept “heater”.
In the example figure 5 the concept Heater is
common to both domain ontologies coming from
two different XML schemas. The pooling of this
concept in two domain ontologies makes it possible
to integrate two XML schema.
Among the languages of representation
developed at the conceptual level there are three
principal types of models: languages containing
frame (M. Kifer et al., 1995), logics of description
(D. Kayser, 1997), and the model of the conceptual
graphs (J. Sowa, 1984). To define our ontology we
chose the language containing frame because the
development related to the Web services which will
be used in our architecture is carried out in the
programming language directed for JAVA objects.
5 CONCLUSION
After a syntactic approach of XML data this paper
presented an ontology based on the semantic
approach of data XML. The study of the semantic
Generic Concept
Generic Relation
Ontology
Generic Ontology
Generic Property
Ontology
Generic Ontology
Domain Ontology
Domain Ontology
Of “Plumbing”
Domain Ontology
Of “Thermic”
Heater
Pipe
Faucet
Chimney
Fan
WEBIST 2006 - INTERNET TECHNOLOGY
36
structure of XML grammars and ontologies has
inspired the construction of classes representing an
integration ontology of data generated by XML
grammars. This cognitive structure has many
advantages of which the most important one is to be
evolutionary as well on the level of the data as of the
definition of the data structures. The field of
ontologies made it possible to achieve the goal
which is an integration structure of XML data.
Nevertheless, the treatment of XML grammars for
the semantic extraction of knowledge realized by the
user in a manual way thanks to the knowledge of
XML schema is not without defects. This extraction
causes problems in the decomposition of the XML
schemas. How to choose the level of granularity by
limiting the problem of the redundancy of
information? This case is rather current in complex
XML schema where tags refer to tags having an
identifier. In this manner a tag can have two
‘fathers’ so that these tag ‘fathers’ compete with
each other (A. Dekhtyar, 2003). In this case if these
links are not factorized the information of the link is
duplicated for each semantic element which are
referenced. These cases are easily detectable and it is
then as easy to help the user to remove this
redundancy of information.
Our method was tested in a static way on a set of
XML grammar schema as well as a set of documents
XML associated with each XML schema. The future
objectives of development are double. On the one
hand we wish to develop a complete system
allowing the integration of XML schema and XML
data in a dynamic manner through a graphic
interface. This tool will include also tools allowing
requests to the system according to our previous
work. In addition, we wish to test this method for the
integration of Web Services. The Web services are
defined using an XML schema defining the contents
of SAOP documents. Consequently, our architecture
would make it possible to carry out semantic
requests on a set of Web Services.
REFERENCES
A. Baxevanis, The Molecular Biology Database
Collection, Nucleic Acids Research, vol 28, n°1, 2000,
p.1-7 http://nar.oupjournals.org/cgi/content/full/27/1/1
D. Benson, I. Karsch-Mizrachi, D. Lipman, J. Ostell, B.
Rapp, D. Wheeler, GenBank, Nucleic Acids Res, vol.
1, n°28, 2000, p. 15-8,
http://www.ncbi.nih.gov/Genbank/
A. Pan, J. Raposo, M. Álvarez, P. Montoto, V. Orjales, J.
Hidalgo, L. Ardao, A. Molano, Á. Viña, The Denodo
Data Integration Platform, VLDB, Hong Kong, China,
2002
D. Draper, A. Y. HaLevy, D. S. Weld, The Nimble XML
Data Integration System, IEEE International
Conference on Data Engineering, April 02 - 06, 2001,
Heidelberg, Germany
M. J. Carey, J. Kiernan, J. Shanmugasundaram, E. J.
Shekita, S. N. Subramanian, XPERANTO :
Middleware for Publishing Object-Relational Data as
XML Documents, The VLDB Journal, pp 646-648,
2000
A. Cali, G. De Giacomo, M. Lenzerini, Models for
Information Integration: Turning Local-as-View into
Global-as-View, Proceedings of the International
Workshop on Foundations of Models for Information
Integration, 2001.
K. Aberer, P. Cudre-Mauroux, M. Hauswirth. A
framework for semantic gossiping. SIGMOD Record,
31(4), 2002
N. Guarino, C. Carrara, P. Giaretta, An ontologie of meta-
level categories, in J. Doyle F. S & Torano P., eds.,
Principles of Knowledge representation and
Reasonning, Morgan-Kauffman, pages 270-280, 1994.
B. Bachimont, Engagement sémantique et engagement
ontologique : conception et réalisation d'ontologie en
ingénierie des connaissances, In Charlet J., Zackland
M., Kessel G. & Bourigault D., eds., Ingénierie des
connaissances : évolution récentes et nouveaux défis,
Eyrolles, pages 305-323, 2000.
N. Guarino, The ontological level, in R. Casati B. S. &
White G., eds, Philosophy and the cognitive sciences,
Hölder-Pichler-Tempsky, 1994.
T. Dechilly, B. Bachimont, Une ontologie pour éditer des
schémas de description audiovisuels, extension pour
l'inférence sur les descriptions, In Actes des journées
francophones d'Ingénierie des Connaissances
(IC'2000), 2000.
B. Amann, Du Partage centralisé de ressources Web
centralisées à l’échange de documents intensionnels,
Documents de Synthèse, 2003.
I. F. Cruz, H. Xiao & F. Hsu, An Ontology-based
Framework for Semantic Interoperability between
XML Sources, In Eighth International Database
Engineering & Applications Symposium (IDEAS
2004), July 2004.
M. Klein, Interpreting XML via an RDF schema. In ECAI
workshop on Semantic Authoring, Annotation &
Knowledge Markup (SAAKM 2002), Lyon, France.
L. V. Lakshmannan, F. Sadri, Interoperability on XML
Data, In Proceeding of the 2nd International Semantic
Web Conference (ICSW’03), 2003.
M. Kifer, G. Laussen, J. Wu, Logical foundations of
object-oriented Land frame-based languages, In
journal of the ACM, 1995.
D. Kayser, La représentation des connaissances, Hermès,
1997.
J. Sowa, Conceptual structures: information processing in
mind and machine, Addison-Wesley, 1984.
A. Dekhtyar, I. E. Iacob, A Framework for Management
of Concurrent XML Markup, International Conference
on Conceptual Modeling, in ER 2003, pages 311-322,
2003.
J. Berstel, L. Boasson, XML Grammars MFCS 2000: 182-
191
ONTOLOGY-BASED INTEGRATION OF XML DATA - Schematic Marks as a Bridge Between Syntax and Semantic
Level
37
