Paper 12 International Journal

Document Sample
Paper 12 International Journal Powered By Docstoc
					                                                                            International Journal of Computer Information Systems,
                                                                                                                Vol. 4, No. 3, 2012

                 Ontology Based Web Service Discovery
                       using Semantic Techniques
  Ilango Paramasivam                Anthoniraj Amalanathan                Ananda Kumar S                      Vijayakumar K
 School of Computing                 School of Computing                School of Computing                School of Computing
Science and Engineering             Science and Engineering           Science and Engineering             Science and Engineering
     VIT University                     VIT University                     VIT University                     VIT University
 Vellore - 632014, India                 Vellore, India                     Vellore, India                     Vellore, India           

Abstract—Web Service discovery becomes an issue of vital            servers, throws great challenge to select the web service which
importance in utilizing available web resources. Current web        suits the user query the most. The proposed web service
service discovery mechanisms rarely take context into               discovery mechanism is based on the standard of UDDI [24]
consideration, leading to poor user experiences. Semantic           [28]. The metadata information can be added to web services
Ontology-based information retrieval is an effective approach to
achieve resource retrieval. In this paper, we propose an approach
                                                                    by using semantic web. Ontology’s can be used to model the
that makes a good use of contextual information from both user      relationship between various entities into real world concepts
query and service advertisement to offer better results. The        which are understandable to machines [16] [20].
proposed approach is Semantic Web based service matching that
complements logic based reasoning with approximate matching              OWL-S is OWL based ontology for encoding properties
based on syntactic information retrieval based similarity           of Web services. It is used to facilitate service annotation
computations. This approach is applied to services and requests     and matching [7][23]. OWL-S ontology defines a service
specified in Semantic Markup for Web Services (OWL-S) [23].         profile for encoding a service description, a service model for
An ontology-based model of context is constructed for semantic      specifying the behavior of a service, and service grounding for
searching; domain ontology’s are used for match process between
request and advertised services. The experiments prove that,
                                                                    how to invoke the service [20] . Figure 1 shows the process of
under certain constraints logic based approaches to OWL-S           ontological ordering and grouping.         Actually, by using
service I/O matching can only be significantly outperformed by      domain ontology [22] described in OWL [4] [20], using
the proposed ones.                                                  special software such as protégé [17], a service searching
                                                                    process involves a matching between the profile of a service
Keywords-Semantic Web, Ontology, Web Service, Information           advertisement and the profile of a service request. The service
Retrieval                                                           profile describes the functional properties such as inputs,
                                                                    outputs, preconditions, and effects, and non-functional
                      I.    INTRODUCTION                            properties such as service name, service category, and aspects
    With the popularization of the Internet and increasingly        related to the quality of service.
diverse online information, many web services are available,
as the number of web services increases, discovery and
selection of appropriate web services becomes more important
than ever. Universal Description Discovery and Integration
(UDDI), is one of the first methods web service publication
and discovery. UDDI [28] provides a common architecture to
advertise and search web services between web services
providers and requesters. The limitation with the UDDI’s
discovery mechanism is that it supports only text based
searching. The search by keywords imposes two constraints:
(i) two semantically equivalent query keywords could not be
identified and (ii) search by keywords is not suitable for web
service discovery. Furthermore, UDDI does not support search
by service capabilities and other properties. This makes UDDI                   Figure 1. Top level of service ontology.
search method being a low precision method. The lack of
semantic support and sufficient discovery precision in UDDI

       March Issue                                          Page 75 of 84                                   ISSN 2229 5208
                                                                          International Journal of Computer Information Systems,
                                                                                                               Vol. 4, No. 3, 2012
    Generally, the semantic service profile provides the           importance in utilizing grid facilities. As the computational
information needed for an agent to discover a service in cross     grid is evolving toward a service oriented computing
platforms. Taken together, the semantic model and discovery        infrastructure, service discovery has been a research focus in
objects associated with a semantic service provide enough          the grid community. Grid information services such as Globus
information for an agent to make use of a service. In this         MDS and R-GMA facilitate discovery of resources and
paper we propose a semantic web based approach for service         services in a grid environment. However, they are restricted to
discovery using ontologies [16].                                   keyword-based queries.
    The rest of this paper is structured as follows: Section II       Quite a few semantic Web service matchmakers have been
discusses the related works in Web Service Discovery, Section      developed in the recent years such as the OWLS UDDI
III describes proposed approach of web service searching           matchmaker, RACER. The majority of them does perform
using ontology’s and matching degree, Section IV describes         profile based service signature (I/O) matching. Alternate
experimental setup, Section V gives performance analysis and       approaches propose service process-model matching, recursive
Section VI concludes this research work with its findings and      tree matching. Except LARKS [18] none of them is
future enhancements.                                               semantically hybrid, in the sense that it exploits both explicit
                                                                   and implicit semantics by complementary means of logic
                                                                   based and approximates matching. LARKS does not perform
                      II. RELATED WORK                             any subsumes and subsumed by nor Nearest-Neighbour
   As the computational grid is evolving toward a service          matching.
oriented computing infrastructure, service discovery has been a       These approaches do not exploit semantics that are implicit.
research focus in the grid community. Grid information             The objective of hybrid semantic Web service matching is to
services such as Globus MDS and R-GMA facilitate discovery         improve semantic service retrieval performance by
of resources and services in a grid environment. However, they     appropriately exploiting means of both crisp logic based and
are restricted to keyword-based queries. UDDI [24] is an           approximate semantic matching where each of them alone
industry initiative for discovery of Web services. UDDI has        would fail.
been utilized by the grid community for discovery of grid             The proposed system also perform profile based service
services. Similar to Globus MDS (Monitoring and Discovery          signature (I/O) matching, however apart from exact match,
System) [2], UDDI only supports keyword matching when
                                                                   plug-in match, subsumed match it also supports subsumed-by
searching for services. Various UDDI extensions have been
                                                                   match and nearest-neighbour match. So, it provides more
proposed to enhance service discovery.
                                                                   generalized service discovery. Hybrid semantic service
   Key success of effectively retrieving relevant services in      matching performed by the matchmaker it exploits both logic-
the future Semantic Web is the degree at which the effective       based reasoning and content-based information retrieval
and accurate intelligent service agents may perform semantic       techniques for OWL-S service profile I/O matching
matching in a way that it goes far beyond of what standard
service discovery protocols can deliver. Central to the
majority of contemporary approaches to semantic service                    III. ONTOLOGY BASED SEMANTIC SEARCHING
matching is that the formal semantics of services specified, for
                                                                       The ontology based Web service searching and handling
example, in OWL-S [7] or WSMO [19] are explicitly defined
                                                                   search queries is shown in Figure 2. Similar to that of a
in some decidable description logic based ontology language        traditional Searching process, a user submits queries in OWLS
such as OWL-DL or F-Logic, respectively. This way, standard        form to the system, wherein the corresponding query specifies
means of description logic reasoning can be exploited to           service profile. Then user selects the searching degree namely
automatically determine services that semantically match with      exact, subsume, subsume-by, nearest-neighbour, system then
a given service request based on the kind of terminological        returns an initial web service set obtained with semantic based
concept subsumption relations computed in the corresponding        retrieval method using domain ontology’s [22]. Finally, the
ontology. Prominent examples of such logic-based only              system computes the similarity of them, and ranks the services
approaches to semantic service discovery are provided by the       according to the level of similarity’s degree before presentation
OWLS-UDDI [24] matchmaker, RACER, and the WSMO                     to the user.
(Web Service Modeling Ontology) [19] service discovery
approach.                                                          A. OWL-S Services
   Various resources on the Internet including processors, disk       In the following, we briefly introduce the essentials of the
storage, network links, instrumentation and visualization          semantic Web service description language OWL-S that are
devices, domain applications, and software libraries can be        needed to understand the concepts of web service matching.
exposed as OGSA/WSRF based grid services, which are                OWL-S [7] is an OWL-based Web service ontology [18],
usually registered with a service registry. A service bus          which supplies a core set of web language constructs for
building on service-oriented grid middleware technologies          describing the properties and capabilities of web services in
such as Globus enables the instantiation of grid services. A       unambiguous and computer-interpretable form. The overall
grid environment may host a large number of services.              ontology consists of three main components: the service
Therefore, service discovery becomes an issue of vital             profile for advertising and discovering services; the process

      March Issue                                          Page 76 of 84                                   ISSN 2229 5208
                                                                            International Journal of Computer Information Systems,
                                                                                                                Vol. 4, No. 3, 2012
model, which gives a description of a service’s operation; and
the grounding, which provides details on how to interoperate                        INPUT(service) = INPUT(request)
with a service, via messages. Specifically, it specifies the                                   AND                                        (1)
signature that is the inputs required by the service and the                       OUTPUT(service)=OUTPUT(request)
outputs generated; furthermore, since a service may require
external conditions to be satisfied, and it has the effect of                      TABLE I. DIFFERENT TYPES OF MATCH
changing such conditions, the profile describes the
                                                                           Different      Definition Informal         Mathematical
preconditions required by the service and the expected effects             Methods                                        Form
that result from the execution of the service.                                          If the inputs and outputs       Services S
    .                                                                                   of R are equivalent          exactly matches
                                                                           Exact        concepts with the inputs       request R ⇔
                                         Semantic                                       and outputs of S              ∀ Is∋ IR : Is- IR^
         User Interface                   Service                                       respectively                 ∀ OR∋ OS : Is- IR
                                                                                        If the outputs of S are
                                                                                                                        Services S
                                                                                        direct subclasses of the
                                                                                                                        plugs into
                                                                                        output of R and the
                                                                                                                      request R ⇔
                                                                           Plug-In      Inputs    of    R    are
                                                                                                                     ∀ Is ∋ IR : Is ? IR^
                                                                                        subsumed by the inputs
                                                                                                                     ∀ OR ∋ OS : Is ℰ
                                                                                        of S in the domain
                                      Service Registry                                                                   LSC(OR)
        Internet or Grid                                                                ontology
         Infrastructure                                                                 If the output of S are
                                                                                                                       Request R
                                                                                        subsumed      by     the
                                                                                        outputs of R and the
                                                                                                                      service S ⇔
                                                                           Subsumes     inputs    of    R    are
                                                                                                                     ∀ Is∋ IR : Is ? IR^
                                                                                        subsumed by the input
                                                                                                                     ∀ OR∋ OS : OR-
                                                                                        of S in the domain
                                         Ontology                                       If the outputs of R are         Request R
                                         Domain                                         direct subclasses of the      subsumed by
                                         Ontology                                       output of S and the            service S ⇔
     Figure 2. Architecture of service discovery system.                   Subsumed     Inputs    of    R    are     ∀ Is ∋ IR : Is ? IR^
                                                                           -By          subsumed by the inputs       ∀ OR ∋ OS : (OS
                                                                                        of S in the domain             = OR V Os ℰ
    The majority of current OWL-S matchmakers perform                                   ontology                       LGC(OR))^
service I/O based profile matching that exploits defined                                                              SimR(S,R) ? α
semantics of concepts as values of service parameters Inputs                            Service S fails to match
and Outputs. There exists no implemented matching algorithm                             with        request     R     Services S is
                                                                           Logic –
that performs an integrated service matching by means of                                according to the above            nearest
                                                                           Based Fail
                                                                                        logic-based      semantic       neighbor of
additional reasoning on logically defined preconditions and                             filter criteria                request R ⇔
effects. Related work on logic based semantic web rule                                  The      weakest    filter     ∀ Is ∋ IR : Is ?
language is on-going.                                                      Nereast-     which can also match                IR ^
                                                                           Neighbour    for disjoint classes or         ∀ OR ∋ OS :
                                                                                        siblings                         OR ? OS V
B. Semantic Matchmaking                                                                 Service S does not            SimR(S,R) ? α
   The semantic service searching performed by the matcher                              match with request R
                                                                                        according to any of the
exploits both logic-based reasoning and content-based
                                                                                        above filters
information retrieval techniques for OWL-S service profile
I/O matching. In the following, we define the semantic                     For all Methods I and O Represents Input and Output
matching degree, the five variants according to the used
similarity metrics.                                                (ii) Plug-in Match
                                                                        Service S plugs into request R that is relaxing the exact
   Let the terminology’s be, LSC: the least specific concept       matching constraint, service S may require less input than it
(direct children), LGC(C) the set of least generic concepts        has been specified in the request R. This guarantees at a
(direct parents).                                                  minimum that S will be executable with the provided input if
                                                                   the involved OWL input concepts can be equivalently mapped
(i) Exact Match                                                    to WSDL [21] [30] input messages and corresponding service
    Service S exactly matches the Request R, that is,              signature data types.

      March Issue                                          Page 77 of 84                                         ISSN 2229 5208
                                                                             International Journal of Computer Information Systems,
                                                                                                                 Vol. 4, No. 3, 2012
               INPUT(service) ≥ INPUT(request)                                         IV. EXPREMENTAL SET-UP
                            AND                              (2)       The ontology based semantic searching concept can be
              OUTPUT(service) ∈ sc(OUTPUT(request))                  implemented by making five modules.
   We assume this as a necessary constraint of each of the                          TABLE II. MODULES DESCRIPTION
subsequent filters. In addition, S is expected to return more
specific output data whose logically defined semantics is                        Modules               Description
exactly the same or very close to what has been requested by
the user. This kind of match is borrowed from the software                     Services        For registering the services
engineering domain, where software components are
considered to plug-in match with each other as defined above                   Requests        For setting the required
but not restricting the output concepts to be direct children of
those of the query.
                                                                               Searching       For setting the required
(iii) Subsume Match                                                            Testing         To apply the algorithm
                Request R subsumes service S, that is                          Results         To display the results

                INPUT(service) ≥ INPUT(request)
                            AND                              (3)         For each service request, the semantic service modeler is
              OUTPUT(request) ≥ OUTPUT(service)                      attached and enhanced ontology match maker algorithm is
                                                                     applied on user query to retrieve the best results. The
This filter is weaker than the plug-in filter with respect to the    Performance and degree of semantic match making results are
extent the returned output is more specific than requested by        identified by applying semantic filters and Information
the user, since it relaxes the constraint of immediate output        Retrieval (IR) similarity metric. Hence this kind of hybrid
concept subsumption. As a consequence, the returned set of           technology will return the results as hypothetically relevant.
relevant services is extended in principle.
                                                                        The proposed system is analyzed by taking the information
(iv)     Subsumed-By Match                                           security ontology which is shown in Figure 3. The user made a
                                                                     request with input as vulnerability repairing, antivirus and
   Request R is subsumed by service S that is                        output as scanning account. Service S1 is considered
                                                                     semantically relevant to request R, since it returns for any
               INPUT(service) ≥ INPUT(request)                       given vulnerability protection and antivirus, the scanning
                             AND                             (4)     account of antivirus. Likewise, service S2 is relevant to R,
            OUTPUT(service) ∈ GC(OUTPUT(request))
                  Similarity(S, R) ≥ threshold

   This filter selects services whose output data is more
general than requested, hence, in this sense, subsumes the
request. We focus on direct parent output concepts to avoid
selecting services returning data which we think may be too

 (v) Nearest-neighbour

       Service S is nearest neighbour of request R that is

                INPUT(service) ≥ INPUT(request)
                               AND                           (5)
               OUTPUT(request) ≥ OUTPUT(service)                                 Figure 3. Information Security Ontology
                   Similarity(S, R) ≥ threshold.                        Since it returns those cleaned registry which are cleared by
                                                                     antivirus both services S1 and S2 should be returned as
                                                                     matching results to the user. However, the logic based only
                                                                     search S1 as plug-in matching with R but fails to return S2.
                                                                     The unfolded concepts of cleaning registry and scanning

          March Issue                                        Page 78 of 84                                   ISSN 2229 5208
                                                                         International Journal of Computer Information Systems,
                                                                                                               Vol. 4, No. 3, 2012
account are same; therefore the S2 is searched as nearest          number of services as the overhead of matching is mainly
neighbour match.                                                   caused by the initialization process of PELLET reasoner.

                  S1:                                                      TABLE III. SAMPLE REQUEST QUERIES AND RESULTS
           IN: Vulnerability
              protection,                    Plugin
               antivirus                                                    Request          Selected Match
                                                                                                                           Actual Result

            OUT: Scanning
              account                                               personmotorcycle_car_
                                            Request R:                                       Exact            Exact        Exact(3)
                                                IN:                 price _service
                                            repairing,                                                        Nearest      Exact(3)
                                             antivirus              personmotorcycle_car_    Nearest
                                                                    price _service           Neighbour
                                                                                                              Or above     Nearest (5)
                                          OUT: Scanning                                                                    Exact(9)
                                            account                                                           Plug-in
                                                                    computer_price_service   Plug-in
                  S2:                                                                                         Or above
           IN: Vulnerability
              protection,                                                                                                  Exact(9)
               antivirus                     Nearest                                                          Subsumes
                                            Neighbour               computer_price_service   Subsumes                      Plug-in(3)
                                                                                                              Or above
            OUT: cleaning
                                                                                                              Exact else
                                                                    book_author_service      Exact                         No match
           Figure 4. Domain Ontology Description                                                              No match
                                                                                             Subsumed -       Subsumed –
   However, if we give same request and advertised services         book_author_service      By               By
                                                                                                                           Subsumed –
to the existing search maker, only S1 will be displayed as                                                    Or above
relevant service to query, but S2 is also relevant to user,
therefore our approach is more novel and efficient.
                                                                      Other overhead involved in service matching is small, e.g.,
                 V.     PERFORMANC ANALYSYS                        the overhead of PELLET in parsing an ontology query is just a
                                                                   few hundred microseconds as it is a light weight reasoner. As
   We collected more than 100 services (retrieved mainly           a result, the overall overhead of matching is not affected by
from IBM UDDI registries and transferred from WSDL to              the number of services.
OWL-S) to measure the performance of our project. This set
of services includes various application domains like
education, travel etc. Then we collected some queries which
are associated with some of the services in the service set.
Then we performed the test for service discovery and we
compared it with existing systems. We found that the number
of discovered documents for a service request is more while
using hybrid semantic matching comparing to logic only based
search. Pure logic based semantic matching by can be
outperformed by hybrid semantic matching, in terms of both
recall and precision.

   That is the case, for example, by use of the best performing
hybrid matchmaker. The main reason is that the additional IR
based similarity check of the nearest-neighbour filter allows us
to find relevant services that pure logic based semantic
matching would fail to retrieve.

   When we compare the results with discovery using UDDI
we observe that UDDI has the least overhead when matching                             Figure 5. Precision and Recall
services as UDDI only supports keyword matching and does
not incur a time-consuming reasoning process. The overhead            Precision and recall are standard measures that have been
of our system does not change much with an increase in the         used in information retrieval for measuring the accuracy of a
                                                                   search method. Recall is the relevant document retrieved and

      March Issue                                          Page 79 of 84                                        ISSN 2229 5208
                                                                                        International Journal of Computer Information Systems,
                                                                                                                            Vol. 4, No. 3, 2012
precision is the accuracy of the document. The best IR
                                                                                 [10] Pengbin Fu1,Shanshan Liu2,Huirong Yang2,Liheng Gu “Matching
similarity metric (Cosine/TFIDF) performs close to the pure                      Algorithm of Web Services Based on Semantic Distance” ISBN 978-952-
logic based search. But pure logic based search is only                          5726-06-0 Proceedings of the 2009 International Workshop on Information
superior to IR based matching at the very expense of its recall.                 Security and Application (IWISA 2009).
However, for discovering semantic web services, precision
                                                                                 [11] Evren Sirin, Bijan Parsia, and James Henler “Filtering and Selecting
may be more important to the user than recall.                                   Semantic Web Services with Interactive Composition Techniques” 1541-
                                                                                 1672/04/ 2004 IEEE.

                           VI.    CONCLUSION                                     [12] Duygu CELiK and Atilla ELÇi “Searching Semantic Web Services: An
                                                                                 Intelligent Agent Approach Using Semantic Enhancement of Client Input
   The main problems of current web service Technologies                         Term(s) and Matchmaking Step” Proceedings of the 2005 International
are the shortage of semantic parts, increasing number of web                     Conference on Computational Intelligence for Modelling, Control and
services in the web, and syntactic-based search operation for                    Automation, and International Conference on Intelligent Agents, Web
web services. These problems make discovering of appropriate                     Technologies and Internet Commerce (CIMCA-IAWTIC’05) IEEE.
web services challenging. Our approach to semantic Web                           [13] W. Johnston, “Semantic Services for Grid-Based, Large-Scale Science,”
service searching, utilizes both logic based reasoning and                       IEEE Intelligent Systems, vol. 19, no. 1, pp. 34-39, Jan./ Feb. 2004.
content based reasoning for semantic Web services. The
system specifies some relationships between those                                [14] D. Martin, “OWL-S 1.1 Release,” 1.1,
input/output parameters such as EXACT, PLUGIN,
SUBSUME, SUBSUME-BY, NEAREST-NEIGHBOUR and                                       [15] Universal Description, Discovery and Integration (UDDI)[EB/OL].
FAIL. According to those relationships the system serves                         (2005-05 20).
appropriate web services to the client based on client request.
                                                                                 [16] Sheila Kinsella, Uldis Boj¯ars, Andreas Harth, John G. Breslin, Stefan
Building semantic Web service matchmakers purely on                              Decker, “An Interactive Map of SemanticWeb Ontology Usage” 12th
description logic reasoners may be insufficient, hence should                    International Conference Information Visualisation IEEE 2008.
give a clear impetus for further studies, research and
development of more powerful approaches to service                               [17] Gao Shu , Omer F. Rana , Nick J. Avis , Chen Dingfang, “Ontology-
                                                                                 based semantic matchmaking approach” Advances in Engineering Software
searching using semantic Web across disciplines.                                 38 (2007) 59–67 science direct.

                                                                                 [18] Min Liu , Weiming Shen , Qi Hao , Junwei Yan, “An weighted ontology-
                              REFERENCES                                         based semantic similarity algorithm for web service” Expert Systems with
                                                                                 Applications 36 (2009) 12480–12490 science direct.
[1] Maozhen Li, Member, IEEE, Bin Yu, Omer Rana, and Zidong Wang,
Senior Member, IEEE ”Grid Service Discovery with Rough Sets” IEEE                [19] Georgios Meditskos and Nick Bassiliades, “Structural and Role-Oriented
transactions on knowledge and data engineering, vol. 20, no. 6, june 2008.       Web Service Discovery with Taxonomies in OWL-S” IEEE
                                                                                 TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, VOL.
[2] Liming Chen, Nigel R. Shadbolt, and Carole A. Goble “A Semantic Web-         22, NO. 2, FEBRUARY 2010.
Based Approach to Knowledge Management for Grid Applications” IEEE
transactions on knowledge and data engineering, vol. 19, no. 2, february 2007.   [20] CHEN Rui, ZENG Zhi-Hao, ZHAO Kai et al, “An Approach to
                                                                                 Construct and Parse the OWL-Based Service Ontology” 2009 International
[3] S. Miles, J. Papay, V. Dialani, M. Luck, K. Decker, T. Payne, and L.         Symposium on Intelligent Ubiquitous Computing and Education.
Moreau, “Personalised Grid Service Discovery,” IEE Proc. Software, special
issue on performance eng., vol. 150, no. 4, pp. 252- 256, 2003.                  [21] Mauricio Espinoza and Eduardo Mena, “Discovering Web Services
[4] D.L. McGuinness and F. van Harmelen, OWL Web Ontology Language               Using Semantic Keywords” 1-4244-0865-2/07 IEEE.
Overview, World Wide Web Consortium (W3C) recommendation,, Feb. 2004.                                    [22] Xiaogang Ji, “Research onWeb Service Discovery Based on Domain
                                                                                 Ontology” ,978-1-4244-4520-2/09/ ©2009 IEEE.
[5] Hrudaya Ku. Tripathy, B.K.Tripathy “A Rough Set Approach for
Clustering the Data Using Knowledge Discovery in World Wide Web for E-           [23] Falak Nawaz , H. Farooq Ahmad , Hiroki Suguri , Arshad Ali, “Semantic
Business“IEEE International Conference on e-Business Engineering                 Web Service Registry for efficient Discovery of OWL-S based Web
                                                                                 Services”, IEEE.
[6] F. Curbera, M. Duftler, R. Khalaf, W. Nagy, N. Mukhi, and S.
Weerawarana, “Unraveling the Web Services: An Introduction to SOAP,              [24] Changjun Hu, Huayu Li, Xiaoming Zhang “A Framework of Web
WSDL, and UDDI,” IEEE Internet Computing, vol. 6, no. 2, pp. 86-93, 2002.        Services Description and Discovery Based on OWL-S and Domain
                                                                                 Ontology”, 2008 IEEE Asia-Pacific Services Computing Conference.
[7] OWL-S. Semantic markup for web services; w3c member submission 22
november    2004.               [25] Nicholas Gibbins, Stephen Harris, Nigel Shadbolt, “Agent-based
20041122/.                                                                       Semantic Web Services”, Web Semantics: Science, Services and Agents on
                                                                                 the World Wide Web 1 (2004) 141–154.
[8] Massimo Paolucci, Takahiro Kawamura, Terry R. Payne, and Katia Sycara
“Semantic Matching of Web Services Capabilities” ISWC 2002, LNCS 2342,           [26] Asunción Gómez-Pérez and Oscar Corcho, “Ontology Languages for the
pp. 333–347, 2002.                                                               Semantic Web” 1094-7167/02/ © 2002 IEEE intelligent systems.

[9] Gongzhen Wang, Donghong Xu, Yong Qi, Di Hou “A Semantic Match                [27] Deborah L. McGuinness and Richard Fikes, James Hendler, Lynn
Algorithm for Web Services Based on Improved Semantic Distance” IEEE             Andrea Stein, “DAML+OIL: An Ontology Language for the Semantic Web”
DOI 10.1109/NWeSP.2008.25.                                                       1094-7167/02/ © 2002 IEEE intelligent systems.

        March Issue                                                     Page 80 of 84                                           ISSN 2229 5208
                                                                                      International Journal of Computer Information Systems,
                                                                                                                          Vol. 4, No. 3, 2012
[28] Organization for the Advancement of Structure Information Standards.
UDDI Executive Overview: Enabling Service Oriented Architecture. /pubs/uddi-exec-wp.pdf. October 2004.

[29] XML xml/default.asp.

[30] Web Services Description Language 2.0,

                         AUTHORS PROFILE
                       Dr. Ilango Paramasivam received his Ph.D
                       Degree in 2009 from National Institute of
                       Technology, Tiruchirappalli - 620015, India. His
                       research interest includes Data Mining &
                       Warehousing, Web Mining, Machine Learning
                       and Information Security.

                       Anthoniraj Amalanathan holds both Bachelor
                       and Master degree in Information Technology;
                       Currently he is doing research in Semantic Web
                       Personalization and Information Retrieval. He is a
                       freelancer and also contributing Open Source
                       Community in software development.

                        Ananda Kumar S received his M.Tech Degree in
                        2007 from VIT University, Vellore - 632014,
                        India. He is pursuing Ph.D in wireless sensor
                        network and Data Mining.

                       Vijayakumar K received his M.Tech Degree in
                       2006 from VIT University, Vellore - 632014,
                       India. He is pursuing Ph.D in Data Mining &

       March Issue                                                    Page 81 of 84                                   ISSN 2229 5208

Shared By: