A Survey on Designing Metrics suite to Asses the Quality of Ontology

Document Sample
A Survey on Designing Metrics suite to Asses the Quality of Ontology Powered By Docstoc
					                                                                (IJCSIS) International Journal of Computer Science and Information Security,
                                                                Vol. 8, No. 8, November 2010

      A Survey on Designing Metrics suite to Asses the
                   Quality of Ontology
                  K.R Uthayan                                                         G.S.Anandha Mala, Professor & Head,
                  Department of Information Technology,                               Department of Computer Science & Engineering
                  SSN College of Engineering                                          St.Joseph’s College of Engineering,
                  Chennai, India                                                      Chennai, India

   Abstract---With the persistent growth of the World Wide Web,                                  II.         LITERATURE SURVEY
the difficulty is increased in the retrieval of relevant information for a            Ahluwalia et al., [1] presented a Semiotic Metrics Suite for
user’s query. Present search engines offer the user with several web
                                                                                   Assessing the Quality of Ontologies. Table 1 shows some of
pages, but different levels of relevancy. To overcome this, the
Semantic Web has been proposed by various authors to retrieve and                  the metrics for quality evaluation [1, 3].
utilize additional semantic information from the web. As the                          As a decisive construct, overall quality (Q) is a subjective
Semantic Web adds importance for sharing knowledge on the internet                 function of its syntactic (S), semantic (E), pragmatic (P), and
this has guide to the development and publishing of several                        social (O) qualities [1] (i.e., Q = b1×S + b2×E + b3×P +
ontologies in different domains. Using the database terminology, it                b4×O). The addition of weight is equal to 1. In the absence of
can be said that the web-ontology of a semantic web system is                      pre-specified weights, the weights are assigned to be equal.
schema of that system. As web ontology is an integral aspect of                       Syntactic Quality (S) evaluates the quality of the ontology
semantic web systems, hence, design quality of a semantic web                      according to the way it is written. Lawfulness is the extent to
system can be deliberated by measuring the quality of its web-
                                                                                   which an ontology language’s rules have been obeyed. Not
ontology. This survey focuses on developing good ontologies. This
survey draws upon semiotic theory to develop a suite of metrics that               every ontology editors have error-checking capabilities;
assess the syntactic, semantic, pragmatic, and social aspects of                   however, without correct syntax, the ontology cannot be read
ontology quality. This research deliberates about the metrics that may             and used. Richness is nothing but the proportion of features in
contribute in developing a high quality semantic web system.                       the ontology language that have been used in ontology (e.g.,
                                                                                   whether it includes terms and axioms, or only terms). Richer
                                                                                   ontologies are more valuable to the user (e.g., agent).
  Keywords--- Quality Metrics, Web ontology, Semiotic Metrics,                        Semantic Quality (E) estimates the meaning of terms in the
Semantic Quality, Domain modularity.                                               ontology library. Three attributes are used here are
                                                                                   interpretability, consistency, and clarity. Interpretability deals
                    I.           INTRODUCTION                                      with the meaning of terms (e.g., classes and properties) in the
                                                                                   ontology. In the real world, the knowledge provided by the
S   EMANTIC Web is nothing but the extension of the
    present web in which the web resources are prepared with
formal semantics about their interpretation for the machines.
                                                                                   ontology can map into meaningful concepts. This is
                                                                                   accomplished by checking that the words used by the ontology
                                                                                   be present in another independent semantic source, such as a
These web resources are combined in the form of web
                                                                                   domain-specific lexical database or a comprehensive, generic
information systems, and their formal semantics are usually
                                                                                   lexical database such as WordNet. Consistency is nothing but
characterized in the form of web-ontologies. By means of the
                                                                                   whether terms having a consistent meaning in the ontology.
database terminology, it can be said that the web-ontology of a
                                                                                   For example, if an ontology claims that X is a subclass of Y,
semantic web system is representation of that system [11].
                                                                                   and that Y is a property of X, then X and Y have incoherent
Design quality of a semantic web system can be calculated by
                                                                                   meanings and are of no semantic value. For example,
computing the quality of its web-ontology because web
                                                                                   ontological terms such as IS-A is often used inconsistently.
ontology is the integral element of semantic web systems [25].
                                                                                   Clarity is the term which determines whether the context of
The main concern is that when the design of a web-ontology is
                                                                                   terms is clear. For example, if ontology claims that class
completed, it is suitable time to assess its quality so that in
                                                                                   “Chair” has the property “Salary,” an agent must know that
case, the design is of low quality, it can be enhanced before its
                                                                                   this illustrate academics, not furniture.
instantiation. This helps in saving of considerable amount of
                                                                                      Pragmatic Quality (P) deals with the ontology’s usefulness
cost and effort for developing high quality semantic web
                                                                                   for users or their agents, irrespective of syntax or semantics.
systems. Metrics are considered as the appropriate tools for
                                                                                   Three criteria are used for determining P. Accuracy is whether
estimating quality. This survey focuses on several metrics for
                                                                                   the claims on ontology makes are ‘true.’ This is very tricky to
web ontology quality evaluation.
                                                                                   determine automatically without a learning mechanism or
                                                                                   truth maintenance system. Currently, a domain expert
                                                                                   evaluates accuracy. The measure of the size of the ontology is

                                                                                                               ISSN 1947-5500
                                                         (IJCSIS) International Journal of Computer Science and Information Security,
                                                         Vol. 8, No. 8, November 2010

called as Comprehensiveness. Larger ontologies are more                    whether the       ontology    satisfies    the    agent’s      specific
probable to be complete representations of their domains, and              requirements.
provide more knowledge to the agent. Relevance indicates

                                               TABLE 1: DETERMINATION OF METRIC VALUES

                      Attributes                                                           Determination
                  Overall Quality (Q)                                            Q = b1.S + b2.E + b3.P + b4.O
                 Syntactic Quality (S)                                                S = bs1.SL + bs2.SR

                                                        Let X be total syntactical rules. Let Xb be total breached rules. Let NS
                   Lawfulness (SL)
                                                          be the number of statements in the ontology. Then SL = Xb / NS.

                                                        Let Y be the total syntactical features available in ontology language.
                    Richness (SR)                            Let Z be the total syntactical features used in this ontology.
                                                                                    Then SR = Z/Y.
                 Semantic Quality (E)                                            E = be1.EI + be2.EC + be3.EA
                                                      Let C be the total number of terms used to define classes and properties
                                                                                    in ontology.
                  Interpretability (EI)
                                                      Let W be the number of terms that have a sense listed in WordNet. Then
                                                                                     EI = W/C.
                                                        Let I = 0. Let C be the number of classes and properties in ontology.
                   Consistency (EC)                    ∀Ci, if meaning in ontology is inconsistent, I+1. Therefore, I = number
                                                                    of terms with inconsistent meaning. Ec = I/C.
                                                         Let Ci = name of class or property in ontology. ∀ Ci, count Ai, (the
                     Clarity (EA)
                                                         number of word senses for that term in WordNet). Then EA = A/C.
                Pragmatic Quality (P)                                           P = bp1.PO + bp2.PU + bp3.PR
                                                        Let C be the total number of classes and properties in ontology. Let V
               Comprehensiveness (PO)
                                                          be the average value for C across entire library. Then PO = C/V.
                                                      Let NS be the number of statements in ontology. Let F be the number of
                    Accuracy (PU)                       false statements. PU = F/NS. Requires evaluation by domain expert
                                                                          and/or truth maintenance system.
                                                      Let NS be the number of statements in the ontology. Let S be the type of
                    Relevance (PR)                     syntax relevant to agent. Let R be the number of statements within NS
                                                                              that use S. PR = R / NS.
                  Social Quality (O)                                                 O = bo1.OT + bo2.OH
                                                       Let an ontology in the library be OA. Let the set of other ontologies in
                                                       the library be L. Let the total number of links from ontologies in L to
                    Authority (OT)
                                                        OA be K. Let the average value for K across ontology library be V.
                                                                                   Then OT = K/V.
                                                       Let the total number of accesses to an ontology be A. Let the average
                     History (OH)
                                                             value for A across ontology library be H. Then OH = A/H.
                    Cohesion (Coh)
                                                                   Where SCC is separate connected components
                      Fullness (F)

                   Readability (Rd)

   For the purpose of evaluation, it needs some knowledge of               the type of information the agent uses by ontology (e.g.,
the agent’s requirements. This metric is coarse as it verifies for         property, subclass, etc), rather than the semantics needed for

                                                                                                        ISSN 1947-5500
                                                        (IJCSIS) International Journal of Computer Science and Information Security,
                                                        Vol. 8, No. 8, November 2010

specific tasks (e.g., the particular subclasses needed to
interpret a user’s specific query).                                          In the above equation, n is total number of sub-domains of
   Social quality (O) imitates the fact that agents and                   web-ontology. Similarly, the OAE metric is officially defined
ontologies exist in communities. The authority of an ontology             as ratio of total number of overlapped axioms (tOAs) to the
is nothing but the number of other ontologies that link to it             total number of domain axioms. It can be written as follows:
(define their terms using its definitions). More authoritative
ontologies indicate that the knowledge they provide is
accurate or useful. The history indicates the number of times
the ontology is accessed. Ontologies are more dependable                                                                                     (2)
when they are with longer histories.
   The cohesion (Coh) of a KB is nothing but the number of
separate connected components (SCC) of the graph
                                                                             In the equation given above, n is total number of sub-
representing the KB.
                                                                          domains of web-ontology. Lastly, the KnE metric is the
   The fullness (F) of a class Ci is defined as the actual number
                                                                          difference of total number of overlapped axioms and the total
of instances that belong to the subtree rooted at Ci (Ci(I))
                                                                          number of isolated axioms. It may be written as follows:
compared to the expected number of instances that belong to
the subtree rooted at Ci (Ci`(I)).
   The readability (Rd) of a class Ci is defined as the total of                                                                             (3)
the number attributes that are comments and the number of                    If the resultant KnE value is positive, then the web-ontology
attributes that are labels the class has.                                 is more knowledge enriched, if it is zero, then the web-
   Amjad et al., [2] provided the Web-Ontology Design
                                                                          ontology is average knowledge enriched, and if it is negative,
Quality Metrics. The author proposes design metrics for web-              then the web-ontology is less knowledge enriched.
ontology [21] by maintaining certain recommended principles
like a metric may reach its highest value for perfect quality for         Characteristics Relevancy metric
excellent case and vice versa that is it may reach its lowest                Characteristics Relevancy (ChR) metric gives us the
level for worst case. It is supposed to be monotonic, clear, and          suggestion about how much a given web-ontology is close to a
intuitive. It must correlate well with human decisions and it             user’s specific necessities and the degree of reusability of the
should be automated if possible. The proposed metrics may                 web-ontology. Formally, it is termed as the ratio of the
give notification about how much knowledge can be derived                 number of relevant attributes (nRAs) in a class to the total
from a given webontology; how much it is relevant to a user’s             number of attributes (TnAs) of that class. It can be written as
specific necessities and how much it is effortless to reuse,              follows:
manage, trace and adapt. The metrics provided by the author
are Knowledge Enriched (KnE), Characteristics Relevancy
(ChR) and Domains modularity (DoM).

Knowledge Enriched metric                                                                                                                    (4)
   The reasoning capability of a web-ontology is determined
by Knowledge Enriched (KnE) metric, and it is based on two
sub-metrics so-called Isolated Axiom Enriched (IAE) metric                   where n in above equation represents the total number of
and Overlapped Axiom Enriched (OAE) metric. There are                     classes in the provided web-ontology. ChR metric reveals the
three parts in this axiom namely, predicate, resource and                 proportion of relevant attributes in the web-ontology, and this
object. If none of these is similar with any other axiom of               number gives insights how much a web-ontology is relevant.
identical domain then that axiom is termed as isolated axiom.
If the two axioms have some similar parts, it is said to be               Domain Modularity metric
overlapped. There may be more than a few transitively                        Domain modularity (DoM) metric denotes the component-
overlapped axioms in any domain. This metric determines the               orientation feature of a web-ontology. This metric specifies
percentage of IAE and OAE, and if the former is greater than              the grouping of knowledge in different components of web-
the later one, then the web-ontology can be regarded as less              ontology. The webontology is best manageable, traceable,
knowledge enriched. IAE is officially defined as the ratio of             reusable and adaptable, if it is designed in components
total number of isolated axioms (tIAs) to the total number of             (subdomains). Formally, the DoM metric is given as the
domain axioms (tDAs).                                                     number of sub-domains (NSD) contained in a webontology.
                                                                          This metric also depends on the coupling and cohesion [25]
                                                                          levels of sub-domains, and it is directly proportional to its
                                                                          cohesion level and inversely proportional to its coupling level.

                                                                                                      ISSN 1947-5500
                                                         (IJCSIS) International Journal of Computer Science and Information Security,
                                                         Vol. 8, No. 8, November 2010

                                                                             Class Richness: The class richness (CR) of a knowledge
                                                                           base is defined as the ratio of the number of classes used in the
                                                             (5)           base (C`) to the number of classes defined in the ontology
                                                                           schema (C).

   In the above equation, DCoh indicates the level of domain
cohesion and DCoup represents the level of coupling among
sub-domains of web-ontology domain. DoM metric is a real                      Average Population: Formally, the average population (P)
number indicating the degree of partial reusability of a given             of classes in a knowledge base is defined as the number of
web-ontology.                                                              instances of the knowledge base (I) to the number of classes
   Samir et al., [3] given the OntoQA: Metric-Based Ontology               defined in the ontology schema (C).
Quality Analysis. The metrics presented can highlight key
characteristics of an ontology schema and also its population
and facilitate users to make an informed judgment easily. The
metrics used by the author here are not 'gold standard'
measures of ontologies. Instead, the metrics are projected to                 Importance: The importance (Imp) of a class Ci is defined
estimate several aspects of ontologies and their potential for             as the number of instances that belong to the subtree rooted at
knowledge representation. Rather than describing ontology as               Ci in the knowledge base (Ci(I)) compared to the total number
merely effective or ineffective, metrics describe a certain                of instances in the knowledge base (I).
aspect of the ontology because, in most cases, the way the
ontology is built is largely dependent on the domain in which
it is designed. The metrics defined here are Schema Metrics
and Instance Metrics. The following are metrics considered by
the author:                                                                   Werner [4] provided a Realism-Based Metric for Quality
                                                                           Assurance in Ontology Matching. There are three levels
  The following are some of Schema Metrics:                                introduced to the methodology for the measurement of quality
                                                                           improvements in single ontologies. These levels are:
   Relationship Richness: The diversity of relations and
                                                                           • Level 1: reality, consisting of both instances and
placement of relations in the ontology is defined by this
                                                                                universals and also the various relations that acquire
metrics. An ontology that has many relations further than                       between them;
class-subclass relations is better than taxonomy with no more              • Level 2: the cognitive representations of this reality
than class-subclass relationships. The relationship richness                    personified in observations and interpretations;
(RR) is defined as the ratio of the number of relationships (P)            • Level 3: the publicly accessible concretizations of the
defined in the schema to the sum of the number of subclasses                    cognitive representations in representational artifacts of a
(SC) plus the number of relationships.                                          range of sorts, of which ontologies are examples.

                                                                              Harith et al., [5] defined the metrics for Ranking
                                                                           Ontologies. In this paper AKTiveRank, a prototype system for
                                                                           ranking ontologies is proposed based on the analysis of their
   Attribute Richness: The attribute richness (AR) is defined as           structures. This paper describes the metrics used in the ranking
the average number of attributes (slots) per class. It is given as         system. The ranking measures used are described below:
the ratio of number attributes for all classes (att) to the number
of classes (C).                                                            Class Match Measure
                                                                              The Class Match Measure (CMM) is intended to estimate
                                                                           the coverage of ontology for the provided search terms.
                                                                           AKTiveRank looks for classes in every ontology that have
  Inheritance Richness: The inheritance richness of the                    labels matching a search term either exactly (class label
schema (IRs) is defined as the average number of subclasses                identical to search term) or partially (class label “contains” the
per class. The number of subclasses (C1) for a class Ci is                 search term).
defined as |HC (C1, Ci)|.
                                                                           Density Measure
                                                                             Density Measure (DEM) is deliberated to approximate the
                                                                           representational-density or information-content of classes and
         The following are some of Instance Metrics:                       accordingly the level of knowledge detail. DEM considers
                                                                           how well the concept is additionally specified (the number of

                                                                                                       ISSN 1947-5500
                                                       (IJCSIS) International Journal of Computer Science and Information Security,
                                                       Vol. 8, No. 8, November 2010

subclasses), the number of attributes related with that concept,            Orme et al., [10] described Coupling Metrics for Ontology-
number of siblings, etc.                                                 Based Systems. XML has grown to be frequent in Internet-
                                                                         based application domains such as business-to-business and
Semantic Similarity Measure                                              business-to-consumer applications. It has moreover produced
   Similarity measures have often been used in information               a basis for service-oriented architectures such as Web services
retrieval systems to afford enhanced ranking for query results.          and the Semantic Web, mainly because ontology data
Ontologies can be analyzed as semantic graphs of concepts                employed in the Semantic Web [16, 17] are stored in XML.
and relations, and hence similarity measures can be applied to           Measuring system coupling is a generally accepted software
explore these conceptual graphs. This helps in resolving                 engineering practice connected with producing high-quality
ambiguities.                                                             software products. In many application domains, coupling can
                                                                         be assessed in ontology-based systems before system
   Henry [7, 23] described a Measurement Ontology                        development by measuring coupling in ontology data. A
Generalizable for Emerging Domain Applications on the                    proposed set of metrics determines coupling of ontology data
Semantic Web. The semantic Web is considered as the next                 in ontology-based systems [22] represented in the Web
generation Web of structured data that are automatically                 Ontology Language (OWL), a derivative of XML.
shared by software agents, which apply definitions and                      Andrew et al., [1] define a semiotic metrics suite for
constraints structured in ontologies to correctly process data           assessing the quality of ontologies. A suite of metrics
from contrasting sources. One aspect needed to develop                   proposed here is to assess the quality of the ontology. The
semantic Web ontologies of emerging domains is creating                  metrics evaluate the syntactic, semantic, pragmatic, and social
ontologies of concepts that are common to those domains.                 aspects of ontology quality according to the semiotic theory.
These general ontologies can be used as building blocks to               The author operationalizes the metrics and employs them in a
develop more domain-specific ontologies. However most                    prototype tool called the Ontology Auditor. A primary
measurement ontologies focus on representing units of                    validation of the Ontology Auditor on the DARPA Agent
measurement and quantities, and not on other measurement                 Markup Language (DAML) library of domain ontologies
concepts such as sampling, mean values, and evaluations of               represents that the metrics are feasible and highlights the wide
quality based on measurements. In this paper, the author                 variation in quality between ontologies in the library. The
elaborates on a measurement ontology that represents all these           contribution of the research is to afford a theory-based
concepts. This paper presents the generality of the ontology,            framework that developers can utilize to develop high quality
and describes how it is developed, used for analysis and                 ontologies and that applications can exploit to choose
validated.                                                               appropriate ontologies for a given task. Zhe et al., [24]
   Fensel et al., [8] provided OIL (Ontology Interchange                 provides some Evaluation Metrics for Ontology Complexity
Language): an ontology infrastructure for the Semantic Web.              and Evolution Analysis.
Initially, Researchers in artificial intelligence motivate the              Ying et al., [12] discusses about semantic web. Presently,
development of ontologies [14] to facilitate knowledge sharing           computers are shifting from single, isolated devices into door
and reuse. Ontologies [15] play a key role in supporting                 points to a worldwide network of information exchange and
information exchange across different networks. A                        business transactions called the World Wide Web (WWW).
prerequisite for such a role lead to the development of a joint          For this cause, support in data, information, and knowledge
standard for specifying and exchanging ontologies. The                   exchange has become a key issue in current computer
authors present OIL which satisfies such standards.                      technology. The achievement of the WWW has made it
   Carlos et al., [9] presented an Ontology-based Metrics                increasingly hard to find, access, present, and maintain the
Computation for Business Process Analysis. Business Process              information required by a wide variety of users. In answer to
Management (BPM) aims to support the whole life-cycle                    this problem, many new research initiatives and commercial
required to deploy and maintain business processes in                    enterprises have been provided to enhance available
organizations. Analyzing business processes have a need of               information with machine processable semantics. This
computing metrics that can facilitate determining the health of          semantic web will offer intelligent access to heterogeneous,
business activities and thus the whole enterprise. However, the          distributed information, enabling software products (agents)
degree of automation currently achieved cannot maintain the              [20] to intervene between user needs and the information
level of reactivity and adaptation demanded by businesses. In            sources available. This paper reviews ongoing research in the
this paper the author argue and show how the use of Semantic             area of the semantic web [19], focusing especially on ontology
Web technologies can enhance to an important extent the level            technology.
of automation for analyzing business processes. The author                  Anthony et al., [13, 18] put forward the Complexity and
presents a domain-independent ontological framework for                  coupling metrics for ontology based information. Ontologies
Business Process Analysis (BPA) with support for                         are greatly used in bioinformatics and genomics to
automatically computing metrics. In particular, a set of                 characterize the structure of living things. This research
ontologies for specifying metrics are defined in this paper. The         focuses on complexity metrics for ontologies. These
domain-independent metrics computation engine is defined                 complexity metrics are obtained from semantic relationships
that can interpret and compute them.                                     in an ontology. These metrics will assist for selecting the best

                                                                                                    ISSN 1947-5500
                                                                (IJCSIS) International Journal of Computer Science and Information Security,
                                                                Vol. 8, No. 8, November 2010

ontologies in several application areas, including                                 [5]    Harith Alani and Christopher Brewster,” Metrics for Ranking
                                                                                          Ontologies”, WWW2006, pp 22–26, 2006.
bioinformatics and genomics.                                                       [6]    Maedche and Steffen Staab"Ontology Learning for the Semantic Web",
   Dimitris N. Kanellopoulos [22] described ODELO: an                                     IEEE Intelligent Systems, Vol 16, pp 72-79, 2001
ontology-driven model for the evaluation of learning                               [7]    Henry M. Kim," A Measurement Ontology Generalizable for Emerging
                                                                                          Domain Applications on the Semantic Web", Journal of Database
ontologies. Trying out or renewing an existing learning
                                                                                          Management, Vol 18, pp 20-42, 2006.
ontology [6] and providing evaluation tools to assess its                          [8]    D. Fensel., F.V. Harmelen, I. Horrocks., D.L. McGuinness, and P.F.
quality are fundamental steps before putting an e-learning                                Patel-Schneider, “Oil: An ontology infrastructure for the semantic web”,
system online. Ontology [21, 23] evaluation is a central task                             IEEE Intelligent Systems, pp 38-45, 2001.
                                                                                   [9]    Carlos Pedrinaci and John Domingue,” Ontology-based Metrics
and it is typically the output of an automatic process. This                              Computation for Business Process Analysis”.
paper put forwards an ontology-driven model, called                                [10]   Orme, Anthony M., Yao, Haining, and Etzkorn, Letha, “Coupling
Ontology-Driven model for the Evaluation of Learning                                      Metrics for Ontology-Based Systems,” IEEE Software, Vol. 23, No. 2,
                                                                                          pp. 102-108, 2006.
Ontologies (ODELO), for the estimation of ontologies                               [11]   Chen, Y., Zhou, L., & Zhang, D. “Ontology-Supported Web Service
representing learning resources with respect to several metrics.                          Composition: An Approach to Service-Oriented Knowledge
Syntax contracts with the proper relations between signs (e.g.,                           Management in Corporate Services.” Journal of Database Management,
                                                                                          17(1),pp 67-84, 2006.
words, phrases, sentences) and the construction of new ones.                       [12]   Ying Ding, Dieter Fensel , Michel Klein and Borys Omelayenko, “The
Social metrics imitate the fact that software agents and                                  semantic web: yet another hip?”, Data & Knowledge Engineering,
ontologies coexist and communicate in communities.                                        Vol.41 n.2-3, pp 205-227, 2002.
                                                                                   [13]   Anthony Mark Orme , Haining Yao and Letha H. Etzkorn , "Complexity
Ontology cohesion metrics indicates the degree of relatedness                             metrics for ontology based information", International Journal of
of ontology classes. ODELO is a deductive valuation model                                 Technology Management, 2009.
that identifies the elements of ontological quality for learning                   [14]   Thomas R. Gruber, “A translation approach to portable ontology
                                                                                          specifications”, Knowledge Acquisition, Vol.5 no.2, pp.199-220, 1993.
ontologies. In this paper the author propose a framework for
                                                                                   [15]   Martin Doerr, Jane Hunter and Carl Lagoze,"Towards a Core Ontology
assessing the quality of learning ontologies that constitute the                          for Information Integration", Journal of Digital Information, Vol 4,
basis for intelligent educational Adaptive Hypermedia (AH)                                No.1, 2003.
systems.                                                                           [16]   Jeff Heflin, James Hendler, “A Portrait of the Semantic Web in Action”,
                                                                                          IEEE Intelligent Systems, Vol.16 no.2, pp.54-59, 2001.
                                                                                   [17]   James Hendler, “Agents and the Semantic Web”, IEEE Intelligent
                    III.          CONCLUSION                                              Systems, Vol.16 n.2, pp.30-37, 2001.
                                                                                   [18]   Anthony M. Orme, Haining Yao, Letha H. Etzkorn, "Coupling Metrics
                                                                                          for Ontology-Based Systems," IEEE Software, Vol. 23, no. 2, pp. 102-
  In this survey, the different suite of metrics for evaluating                           108, 2006.
ontologies based upon the semiotic-web is analyzed. It is                          [19]   Alexander Maedche, Steffen Staab, “Ontology Learning for the
                                                                                          Semantic Web”, IEEE Intelligent Systems, Vol.16 no.2, pp.72-79, 2001.
better to assess the quality of web-ontology after the design is                   [20]   Larry M. Stephens, Michael N. Huhns, “Consensus Ontologies:
completed. This helps in neglecting the low quality ontology                              Reconciling the Semantics of Web Pages and Agents”, IEEE Internet
before its development by enhancing those defects in                                      Computing, Vol.5 n.5, pp.92-95, 2001.
                                                                                   [21]   Vijayan Sugumaran , Veda C. Storey, “Ontologies for conceptual
ontology. This helps in saving of considerable amount of cost                             modeling: their creation, use, and management, Data & Knowledge
and effort [2] for developing high quality semantic web                                   Engineering”, Vol.42 no.3, pp.251-271, 2002.
systems. Several metrics such as Knowledge Enriched metric,                        [22]   Dimitris N. Kanellopoulos , "ODELO: an ontology-driven model for the
                                                                                          evaluation of learning ontologies",International Journal of Learning
Characteristics Relevancy metric, Domain Modularity metric,                               Technology archive ,Vol 4, pp 73-99, 2009.
Richness, Instance Metrics, Semantic Similarity Measure,                           [23]   Henry M. Kim, Arijit Sengupta and Joerg Evermann, “MOQ: Web
Density Measure, etc., are analyzed for assessing the quality                             services ontologies for QoS and general quality evaluations”.
of ontology. This survey helps in choosing the best suited                         [24]   Zhe Yang, Dalu Zhang and Chuan YE, “Evaluation Metrics for
                                                                                          Ontology Complexity and Evolution Analysis”, ICEBE '06. IEEE
metrics for assessing the quality of ontology. The different                              International Conference, 2006.
metrics which involve different criteria for ontology are                          [25]   Yinglong Maa, Beihong Jinb and Yulin Fengb, “Semantic oriented
analyzed in this survey.                                                                  ontology cohesion metrics for ontology-based systems”, Journal of
                                                                                          Systems and Software archive, Vol 83, pp 143-152, 2010.

[1]   Punit Ahluwalia, Andrew Burton-Jones, Veda C. Storey and Vijayan
      Sugumaran “A Semiotic Metrics Suite for Assessing the Quality of
      Ontologies”, journal of Data & Knowledge Engineering,vol 55, pp 84-
      102, 2005.
[2]   Amjad Farooq, Syed Ahsan, and Abad Shah, “Web-Ontology Design
      Quality Metrics”, Journal of American Science, 2010.
[3]   Samir Tartir, I. Budak Arpinar, Michael Moore, Amit P. Sheth and
      Boanerges Aleman-Meza, “OntoQA: Metric-Based Ontology Quality
      Analysis”, IEEE Workshop on Knowledge Acquisition from Distributed,
      Autonomous, Semantically Heterogeneous Data and Knowledge
      Sources, 2005.
[4]   Werner Ceusters,” Towards a Realism-Based Metric for Quality
      Assurance in Ontology Matching”, Proceeding of the 2006 conference
      on Formal Ontology in Information Systems: Proceedings of the Fourth
      International Conference (FOIS ) ,pp 321-332, 2006.

                                                                                                                      ISSN 1947-5500