SDRule Markup Language: Towards Modeling and Interchanging Ontological Commitments for Semantic Decision Making

SDRule Markup Language: Towards Modeling and Interchanging Ontological Commitments for Semantic Decision Making

Yan Tang (Free University of Brussels, Belgium) and Robert Meersman (Free University of Brussels, Belgium)
DOI: 10.4018/978-1-60566-402-6.ch005
OnDemand PDF Download:
$30.00
List Price: $37.50

Abstract

The emergence of ontology based applications, e.g. the Semantic Web, marks the importance of ontologies. Application rules, such as decision making rules, are often committed to an existing domain ontology when a new application needs to be designed and developed. During this process, the semantics of application rules is required to be precisely grounded. In this chapter, we tackle the problems of modeling and interchanging ontological commitments in order to support ontology based decision making. We model and visualize ontological commitments by means of an extension to Object Role Modeling Language (ORM), which was called ORM Plus (ORM+) and is now named Semantic Decision Rule Language (SDRule-L). SDRule-L is a commitment language for modeling dynamic and non-monotonic decision rules. SDRule-L models are further stored in an XML-based markup language called Semantic Rule Markup Language (SDRule ML), which is a hybrid language of Rule Markup Language (Rule-ML) and Object Role Modeling Markup Language (ORM-ML). We also illustrate its supporting tool called SDRule-Lex, which is based on Tiny Lexon Browser (T-Lex). We demonstrate in the field of on-line customer management.
Chapter Preview
Top

Introduction And Motivation

An ontology is a semiotic representation of agreed conceptualization in a subject domain (Gruber, 1993; Guarino, 1997). In 1994, Tom Gruber proposed to use relational database schemata as ontologies when he gave the definition of ontology.

“…In the knowledge sharing context, ontologies are specified in the form of definitions of representational vocabulary. A very simple case would be a type hierarchy, specifying classes and their subsumption relationships. Relational database schemata also serve as ontologies by specifying the relations that can exist in some shared database and the integrity constraints that must hold for them.” (Tom Gruber, 1994; SRKB Mailing list, cited by NicolaGuarino, (1997))

In the later 90’s of the last century, Prof. Robert Meersman from VUB STARLab brought forward the idea of applying the principles of database engineering to ontology engineering. The idea later laid the foundation of a framework called Developing Ontology-Grounded Methods and Applications1 (DOGMA, Meersman, 1999 a; Meersman, 1999 b; Meersman, 2001; Spyns et al., 2002), which is designed and inspired by the tried-and-tested principles from conceptual database modeling.

In DOGMA, formally committing an application (e.g. application rules, task processes and application symbols) to a domain ontology is complicated. In order to do this, Object Role Modeling (ORM, Halpin, 2001) is adopted for modeling, validating and visualizing the ontological commitments. In (Demey et al., 2002; Spyns et al., 2002), the authors studied many advantages of using ORM as a commitment modeling language.

Later, Demey et al. (2002) present an XML-based ORM markup language (ORM-ML), which enables exchanging ORM models including ORM application rules. The ORM-ML can be fully mapped to OWL2 (Mustafa, 2007), which makes it possible to adapt many available ontology technologies.

However, ORM still lacks several logical operators and connectors while grounding the semantics for dynamic decision rules, e.g. the sequences and dependences. Moreover, ORM is limited on the use of some specific operators, such as the implication operator. Therefore, we recently propose to design an extension to ORM, the result of which was called ORM+ (Tang et al., 2007; Tang & Trog, 2008), and is now named SDRule-L. We use SDRule-L specifically for modeling semantically rich decision rules, intending to model the ontological commitments for collaborative decision support systems.

Top

Background

This section introduces the background of DOGMA approach to ontology engineering and the ORM approach to ontological commitments. In the meanwhile, we discuss our related work.

Key Terms in this Chapter

DOGMA: DOGMA (Developing Ontology-Grounded Methods and Applications), which is a methodological framework for ontology engineering, was brought forward by Prof. Dr. R. Meersman (1999) and became mature in (Spyns et al., 2002, Jarrar and Meersman, 2007). It has its roots in database semantics and model theory. In DOGMA, an ontology is modeled by the double articulation principle into two layers: the lexon base layer and the commitment layer. A lexon is a plausible binary fact. The lexon base of an ontology describes a universe of discourse (UoD). Lexons are a more formal but still linguistically determined representation of propositions about a domain to be modeled (Spyns et al., 2007). A lexon contains five elements: context identifier, two terms and two roles.

RuleML: RuleML (Rule Markup Language, http://www.ruleml.org) connects rule based expert systems together with knowledge markup technologies. It permits both forward (bottom-up) and backward (top-down) rules in XML for deduction, rewriting, and further inferential-transformational tasks. The RuleML Initiative, which combines rules in natural language and formal notation, is working towards an XML-based markup language that permits Web-based rule storage, interchange, retrieval, and firing/application (Boley et al., 2001, Wagner et al., 2004).

Commitment: Gruber (1993) defines ontological commitments as “agreements to use the shared vocabulary in a coherent and consistent manner”. Guarino et al. (1994) defines ontological commitments as “a map between a logical language and a set of semantic structures”. By using the ontological commitments, an application commits its local vocabulary and application semiotics to the meaning of the ontology vocabulary. A commitment can contain various constraints, rules and axiomatized binary facts based on the needs of applications.

Context: Contexts have been introduced in the ontology base as an organizing principle, grouping related lexons. A context can be considered as an identifier of a possible world, leading to specific possible world semantics and mappings between them (Spyns et al., 2007, Sowa, 2000).

ORM-ML: ORM Markup Language, which has been developed by VUB STARLab in 2002 (Demey et al., 2002, Jarrar et al., 2003), is an XML-based ORM markup language with a complete grammar defined in an XML Schema. It enables exchanging ORM models amongst agents. In order to facilitate validation, e.g. providing formal and consistent documentation, VUB STARLab has also developed a verbalization style sheet for ORM-ML documents that allows presenting the facts and the rules in pseudo natural language sentences.

Ontology: There exist many definitions of ontology, amongst which the definition given by Tom Gruber is a widely accepted one. An ontology is “an explicit specification of a conceptualization …When the knowledge of a domain is represented in a declarative formalism, the set of objects that can be represented is called the universe of discourse. Those sets of objects, and the describable relationship among them, are reflected in the representational vocabulary with which a knowledge-base program represents knowledge…” (Gruber, 1993).

ORM: Object Role Modeling (Halpin, 2001) is a powerful method for designing and querying database models at the conceptual level, where the application is described in terms easily understood by non-technical users (http://www.orm.net/). It has roots in natural language information analysis method (NIAM) illustrated in (Wintraecken, 1990).

Complete Chapter List

Search this Book:
Reset