PROCESSING AND MANAGING COMPLEX OWL DATA

Hussam Abuazab

and Atilla Elci

Department of Computer Engineering, Eastern Mediterranean University, Gazimagusa, North Cyprus, Turkey

Al Ghurair University, Dubai, U.A.E.

Keywords: OWL class sets, DeMorgan’s Law, minusOf, Auxiliary identity.

Abstract: In this paper we introduce new OWL class sets that can be obtained through and manipulated by Boolean

operators. The need for new class sets come in order to process and manage complex OWL data, and to

improve the expressive power of OWL language. A particular need for such sets arises in processing

complex data involved in ontology for human disease which is being developed. The class sets introduced in

this paper include the definition of minusOf, De Morgan's Law, and Auxiliary Identity properties.

1 INTRODUCTION

OWL is rich with constructs in class expression

category like owl:oneOf, owl:disjointWith,

owl:unionOf, owl:complementOf; and in Enumerated

values like owl:dataRange. Also, Boolean class

combinations extended to include owl:intersectOf,

which play major role in representing intricacies in

ontology (Bennacer, 2004). Yet, ontology of

diseases (Ashburner, 2000) give way to complex

data and combinations there of which is very

cumbersome to represent using the existing class

definitions and operators, which result on poor

representations of the ontology. Although there are

several reasons for poor ontology described on

(Hepp, 2006), lack of some class set definition is a

reason for this problem. The new class set will help

on minimize the gap between conceptual data

schemes and ontologies, which rise as hot topic on

the march of developing tools for building strong

ontologies (Jarrar, 2003).

In the following section, we introduce new class sets

to use in representing complex data of ontology of

diseases. Section 3 discusses the use of these new

sets in diagnosing in the presence of symptoms of

multiple ailments. Section 4 concludes the paper.

2 COMPLEX CLASS SETS

In this section we introduce three new OWL class

selectors and their resultant class sets, namely,

minusOf, De Morgan's Law, and Auxiliary

Identity.

2.1 minusOf Class

The minusOf represents the set

BxAxBA

iix

∉

∈

∀

−

&}{ where A & B are

OWL class sets, for example symptoms of two

diseases.

Figure 1: The class of minusOf.

A: symptoms of Disease A.

B: symptoms of Disease B.

So, minusOf (A,B) represents the symptoms of

disease A excluding those shared with the symptoms

of disease B.

237

Abuazab H. and Elci A. (2008).

PROCESSING AND MANAGING COMPLEX OWL DATA.

In Proceedings of the Fourth International Conference on Web Information Systems and Technologies, pages 237-240

DOI: 10.5220/0001528402370240

 SciTePress

2.2 DeMorgan's Law Class

De Morgan's Law class represents the

CBA ∩∩ =

___

CBA ∪∪ , which implies that the

intersect elements are not satisfied. Let’s assume

that A, B, and C are classes of symptoms of three

different diseases, and then

CBA ∩∩ implies that

the common symptoms are not satisfied.

In the way of an example let’s consider the

following OWL description of three diseases on

figure 2, for example, acute sinusitis, pneumonia and

common cold. Each disease has known symptoms,

and some symptoms may be common among all or

between any two diseases.

Figure 2: OWL description of three diseases.

Figure 2 (cont): OWL description of three diseases.

Studying the above OWL description, it is clear that

there are some common symptoms between the three

diseases. Figure 3 shows the intersections between

the three classes of symptoms of the three diseases.

Figure 3: Class of De Morgan's Law.

Bad Breath is the common symptom among the

three diseases. The patient must suffer from bad

breath so he/she may be said to have been infected

by at least one or more of the three diseases, acute

sinusitis, and pneumonia and common cold. In case

the patient is not suffering from bad breath, this

implies that he/she must not have been infected by

any of the three diseases.

This can be represented by De Morgan's Law as

following:

CBA ∩∩ =

___

CBA ∪∪

where A is for symptoms of acute sinusitis, B is for

pneumonia, and C is for common cold.

WEBIST 2008 - International Conference on Web Information Systems and Technologies

238

So instead of calculating the values of

B , and

C , and then finding the union of these classes,

applying De Morgan's Law, will result in

professional solution in means of time and

complexity.

2.3 Auxiliary Classes

Auxiliary Classes represent the sets:

By supposing A: symptoms of Disease A, and B:

symptoms of Disease B, the set

AABA

implies that the patient suffers from symptoms of

disease A, and the shared symptoms of diseases A

and B, so the patient infected by disease A.

AA +=+

−

In this case, the patient suffers from symptoms of

disease A, and symptoms of disease B, except the

ones common with disease A, so the patient effected

by disease A and B.

BCACABA

=++ ))((

In this case, the patient suffers from symptoms of

disease A, and symptoms of disease B, so, let us say,

the patient suffers from sneezing – symptom of

common cold – and from pain and swelling around

eyes – symptom of Acute Sinusitis – also, the patient

suffering from chest pain - symptom of Pneumonia –

so, for sure, the patient is suffering from bad breath

which is the intersect symptom of both Acute

Sinusitis and Pneumonia. See Fig. 2.

3 DISCUSSION

The mentioned three complex classes are comply to

the standards of OWL classes regarding the name

(URIref) with the “rdf:ID” attribute; also, the class

may have properties and instances. The tag for

defining the new classes is OWL:Class.

The

minusOf class is used to implement the cases

where the shared symptoms of two diseases are not

available while the non-shared symptoms of one

disease are available; let us assume a patient is

suffering from sneezing, which may imply that

he/she is infected by common cold, but the patient is

neither suffering from nasal congestion nor bad

breath, which implies the patient is not infected by

acute sinusitis for sure, as A-B is true for the patient.

So the symptoms of pain and swelling around eyes

need not be investigated, and the result is diagnosed

more precisely and in lesser time. Reference Figure

Figure 4: minusOf Class.

The De Morgan's class is used to simplify the

diagnosing path, for example, if the patient is not

suffering from bad breath then he/she is not infected

by either of common cold, acute sinusitis, or

pneumonia. This also indicates that one of the other

non-common symptoms may be investigated to

determine which ailment is present. Reference

Figure 5.

Figure 5: .

The minusOf Class may have the property

hisExcludedInstance to represent the instances of

PROCESSING AND MANAGING COMPLEX OWL DATA

239

class A which are shared with class B. The

DeMorgan'sLaw Class may have the property

nonCommonInstance to represent the complement of

the common instances between the input classes.

Of the

Auxiliary Classes, the first class

=+ , if the patient is suffering from

sneezing – symptom of common cold – and from

nasal congestion – symptom of both common cold

and acute sinusitis – then the patient is infected by

common cold for sure, while more investigations are

required to decide whether he/she infected by acute

sinusitis as well. The second class of the Auxiliary

Classes

−

, states that if the

patient is suffering from sneezing – symptom of

common cold – and from pain and swelling around

eyes – symptom of acute sinusitis but not of

common cold – then the patient is infected either by

common cold or acute sinusitis or both diseases.

The third

Auxiliary Class,

BCACABA +=++ ))((

, indicates that if the

patient is suffering from common cold – disease A –

and sometimes suffering from acute sinusitis or

pneumonia, so the patient for sure suffering from

bad breath.

4 CONCLUSIONS

In this paper, we introduced new ontology classes.

The minusOf class is used for example to represent

the symptoms of a disease except those shared with

some other disease which implies that the patient is

infected by one but not the other disease. The

DeMorgan’s Law class is utilized to check the

availability and non-availability of common

symptoms of more than one disease, which implies

that the patient is not infected by all diseases causing

the shared non-available symptoms. Finally, three

auxiliary classes are defined to represent various

symptoms’ availability.

The

minusOf and DeMorgan's Law classes

introduced in this paper will play prime role on

easing the representation of complex knowledge,

especially medical knowledge, and the knowledge

that requires such sort of complex classes. The three

auxiliary classes defined are meant to ease the

querying process by deducing the common instances

and individuals of different classes.

In the future more of such specialized classes may

be introduced in order to enhance OWL’s level to

represent more complex knowledge. Also, similarity

may be drawn with implementing combination of

logic gates that subsume certain knowledge,

consequently easing the processing of related data

sets.

REFERENCES

Bennacer, N. Bourda, Y. Doan, B. "Formalizing for

querying learning objects using OWL", Supelec,

France Sept. 2004; ISBN: 0-7695-2181-9

M. Ashburner, C. A. Ball, J. A. Blake, D. Botstein, H.

Butler, J. M. Cherry, A. P.Davis, K. Dolinski, S. S.

Dwight, J. T. Eppig, M. A. Harris, D. P. Hill, L. Issel-

Tarver, A. Kasarskis, S. Lewis, J. C. Matese, J. E.

Richardson, M. Ringwald, G. M.Rubin, and G.

Sherlock. Gene ontology: tool for the unification of

biology. the gene ontology consortium. Nat Genet,

25(1):25–9, 2000. 1061-4036 Journal Article.

Martin Hepp, Jos de Bruijn GenTax: A Generic

Methodology for Deriving OWL and RDF-S

Ontologies from Hierarchical Classifications,Thesauri,

and Inconsistent Taxonomies Int’l Journal on

Semantic Web and Information Systems (IJSWIS), ,

pp. (2006b)

Jarrar, M., Demey, J., Meersman, R.: On Using

Conceptual Data Modeling for Ontology Engineering.

In Spaccapietra, S., March, S., Aberer, K., (Eds.):

Journal on Data Semantics (Special issue on Best

papers from the ER, ODBASE, and COOPIS 2002

Conferences). LNCS, Vol. 2800, Springer, pp.:185-

207. October 2003.

WEBIST 2008 - International Conference on Web Information Systems and Technologies

240