A. Borgida. Description logics in data management. IEEE transactions on knowledge and data engineering, 7(1), 1995.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....pointing links (Section 7) argues for the deployment of LDAP instead of the emerging XML technologies in a middleware in our system. Although extensive work has been done in the area of modeling ontologies for data integration, most researchers use logic based languages like description logics [6], or FLogic [23] to describe them and bene t from the reasoning abilities of logic systems. Our approach, on the other hand, uses a relatively simple model with a clear syntax that allows us to provide a uni ed formal framework to be used for both, ontology de nition and information integration ....

A. Borgida. Description logics in data management. IEEE Transactions on Knowledge and Data Engineering, 7(5):671-682, 1995.

....from two generic instruments defined with the meta model: query classes and matrix based visualizations. Query classes [Staudt et al. 1994] are parameterized and possibly materialized views defined by necessary and sufficient membership constraints similar to description logics [Borgida, 1995]. They defer the checking of consistency and completeness of the perspectives generated by parallel development to a moment definable by the teams themselves. Usually, a large number of query classes can be associated with a meta model and later applied in the negotiations between the modeling ....

A. Borgida. Description Logics in Data Management. IEEE Transactions on Knowledge and Data Engineering 7, 5 (1995), pp. 671 - 682.

....usually versioned, making techniques from temporal DBs of interest. Other architectures that combine OO and deductive capabilities in sophisticated ways can also provide significant benefits to model management such as, Te los [MBJK90] ConceptBase [JJ89] F Logic [KLW95] and Description Logics [Bor95] Inferencing in model management: Several key operations in model management involve various forms of inference, such as inverting a mapping, completing a mapping, and determining equivalence of models. For example, a mapping can be thought of as a view of one model in terms of another. ....

Alex Borgida. Description logics in data management. IEEE Transactions on Knowledge and Data Engineering, 7(5):671--682, 1995.

.... [13,16] and MIX [5] The second is based on the local as view approach where each source is described independently as a local view on the mediator domain model which captures the basic vocabulary of a certain domain, expressed in some database or knowledge base formalism (e.g. Description Logics [7]) Information Manifold [3] SIMS [12] PICSEL [26] Infomaster [25] and DISCO [49] are basic examples of such systems. An important part of the research has been directed towards the use of ontologies as mediator domain models. One of the first systems to follow this approach was Carnot [17] ....

A. Borgida. Description Logics in Data Management. IEEE Transactions on Knowledge and Data Engineering, 7(5):671-- 682, 1995.

....cooperative information system (CIS) described in [6] A CIS is quite similar to a federated databases. It is composed of a set of database schemata and a set of so called interschema assertions. Database schemata represent the individual information sources and are theories in description logics [4]. Interschema assertions formalize relations between different database schemata. CISs formalize a certain degree of autonomy, each database having its own language, domain, and schema. Furthermore CISs formalize a certain degree of redundancy by means of interschema assertions, which capture four ....

A. Borgida. Description Logics in Data Management. IEEE Transactions on Knowledge and Data Engineering, October 1995.

....information system (CIS) described in [4] A CIS is quite similar to an information integration state. It is composed of a set of database schemata and a set of so called interschema assertions. Database schemata represent the individual information sources and are theories in description logics [3]. Interschema assertions formalize relations between different database schemata. CISs formalize a certain degree of autonomy, each database having its own language, domain, and schema. Furthermore CISs formalize a certain degree of redundancy by means of interschema assertions, which capture four ....

A. Borgida. Description Logics in Data Management. IEEE Transactions on Knowledge and Data Engineering, October 1995.

.... Description Logic models have been used in a wide variety of data oriented applications including the representation of complex schemas for the car industry [136] and software management [47] A Description Logic model can be seen as a form of expressive compositional objectbased schema [18]. Recent work concentrates on unifying DLs with conventional Database 1.2. BACKGROUND ON DESCRIPTION LOGICS AND GRAIL 23 Management System (DBMS) 118, 127] combining the expressive modelling and querying power of DLs with the efficient management of large data sets and query optimisation of ....

BORGIDA, A. Description Logics in Data Management. IEEE Transaction on Knowledge and Data Engineering 7, 5 (1995), 671--682.

.... Layer 3 Query execution declarative query ordered execution plan Source 1 Source 2 Source 3 (wrapped sources) Logics, are a family of logics explicitly designed to represent taxonomic and conceptual knowledge of an application domain on an abstract level; for an overview see (Borgida 1995). DLs are usually given a Tarski style declarative semantics, which allows them to be seen as sub languages of first order predicate logic. In the TAMBIS project we use the GRAIL DL (Rector 1996) developed at Manchester. Briefly, a DL is an isa based classification system that allows a ....

Borgida A., Description Logics in Data Management. IEEE Transactions on Knowledge and Data Engineering, 7(5): 671-682, 1995.

....As already mentioned, a large number of information integration systems that fall into the second approach rely on a knowledge base formalism to represent the mediator domain model and to describe resources as views over this domain model. The most commonly used formalism is Description Logics [18] that support modeling and reasoning efficiently about data and knowledge in a domain of discourse. For this purpose DL languages provide a set of of constructs to denote concepts and roles and a set of constructors that permit the definition of more complex concepts. Description logic languages ....

....information, to represent ontologies and to formulate queries. Ontobroker query formalism and document annotation languages are oriented towards a frame based representation of ontologies. Ontobroker also allows the specification of inference rules (similar to those in Description Logics [18]) that allow to derive facts not present in the actual documents. Annotations of HTML pages in Ontobroker are done as in SHOE, incorporating special tags in HTML pages. Basically, Ontobroker allows to describe each resource by a URL, considered as an instance of an ontology class, to set the ....

A. Borgida. Description Logics in Data Management. IEEE Transactions on Knowledge and Data Engineering, 7(5):671--682, 1995.

....by a set of local models, and communication between agents (inter schema constraints) is modeled by imposing a compatibility relation between local models. LMS is similar to Catarci and Lenzerini s approach. In [ Catarci and Lenzerini, 1993 ] they formalize DB schemata by description logics [ Borgida, 1995 ] and inter schema constraints by relations between concepts of different schemata. On the one hand, their semantics formalizes a certain degree of distribution, autonomy, and redundancy, as each DB has its own language, domain, and schema; furthermore redundancy is formalized by inter schema ....

A. Borgida. Description Logics in Data Management. IEEE Transactions on Knowledge and Data Engineering, October 1995.

....data (i.e. accuracy, consistency, completeness, etc. Unfortunately, most of the time, Z39.50 wrappers are developed using some programming language and they do not provide abstract mapping languages with verifiable properties [44, 11, 45] In this paper, we advocate a Description Logic framework [9] (such as proposed in the context of the DARPA KSE [41] for the declarative specification of Z39.50 wrappers using high level concept languages. We claim that modeling the required mappings as first class citizens, instead of hard coding them in the wrappers (i) allows the formal validation of ....

A. Borgida. Description Logics in Data Management. IEEE Transactions on Knowledge and Data Engineering, 7(5):671--682, October 1995.

....Logics. These have enjoyed considerable attention from the Knowledge Representation section of the AI community in the last few years and there are a number of well known prototype DLs, descendants of the KL ONE language, including CANDIDE, BACK, CLASSIC, and LOOM; for an overview see [5]. All DLs define complex entities in terms of composite descriptions made up of a limited set of elementary concepts assembled according to explicit rules. DLs can be viewed as languages obtained by term composition using recursive term constructors, where some terms are concepts (denoting a ....

....SportsCar which h hasCountryOfOrigin Germany, hasWheelType Alloy i. Figure 2. A Description Logic Model in GRAIL 2.1. Benefits of a Description Logic Description logics have two functions that make them particularly attractive as models for describing semistructured and complex information [5]. A type system. Expressivity: Description Logics make it possible to express the semantics of information systems and are often more expressive than traditional Semantic Data Models or Object Oriented data models; Type checking: type refinement is provided automatically through the subsumption ....

[Article contains additional citation context not shown here]

A. Borgida. Description Logics in Data Management. IEEE Transactions on Knowledge and Data Engineering, 7(5):671--782, 1995.

....work and extends it by using so called ABox patterns for describing n ary queries. 2 The Description Logic ALCRP(D) This section gives a brief introduction to the description logic ALCRP(D) and to description logics (DLs) in general summarizing the notions important for this paper. We refer to [1, 14] for more information about description logics. Many DLs can be viewed as subsets of first order predicate logic. However it is important to note that particular DLs are only considered as practical if they are based on sound, complete and terminating reasoning algorithms, i.e. the decidability ....

....and reduce the search space to the set of query matches already computed for the query concept from ABox A 1 . Note that these query matches are members of the concept cottage c 1 . This type of query optimization is an important aspect in applying description logics to database theory (see [1] for an introduction to these topics) The benefits of computing a concept subsumption taxonomy can be even more subtle. Imagine a query from another user looking for a cottage located in a forest that is connected to a river. The ABox A 4 is derived from the sketch. The abstraction process ....

A. Borgida. Description logics in data management. IEEE Transactions on Knowledge and Data Engineering, 7(5):671--682, 1995.

....present paper does not include the full blown logical tool resulting from the former stream of research, but rather focuses on the results of the latter stream. A preliminary version of this model can be found in [46] 4 A fuzzy description logic Description Logics (DLs, for short see e.g. [9, 21]) are contractions of the predicate calculus that descend from the formalization of early semantic network or frame based knowledge representation languages. DLs have an object oriented character that makes them especially suitable for reasoning about hierarchies of structured objects. DL ....

A. Borgida. Description logics in data management. IEEE Transactions on Knowledge and Data Engineering, 7:671--682, 1995.

....FRV96] Even in the relational case these works need to be extended for the problems faced in model management, not to mention extensions to other contexts. Description Logics provide another formalism that has been shown to be useful for modeling DB schemas and interschema constraints [CL93, Bor95, LRO96] They provide a formalism for representing and reasoning about intentionally defined sets. Finally, there is a rich literature on tools for testing equivalence and containment of queries [Ull97] starting from the work of Chandra and Merlin [CM77] Efficient operations on models: A model ....

Alex Borgida. Description logics in data management. IEEE Transactions on Knowledge and Data Engineering, 7(5):671--682, 1995.

....Logics (DLs) are a family of knowledge representation languages that allow reasoning with compositional structured information. In particular, a DL supports hierarchical classification through the use of a well defined notion of subsumption. For a full description of DLs and their uses, see [5]. DLs are closely related to propositional modal and dynamic logics recent work has provided a sound formal basis for several DLs along with results concerning their complexity and expressiveness [6, 18] A DL models an application domain in terms of concepts (classes) roles (relations) and ....

....be asserted to be instances of particular concepts and pairs of individuals can be asserted to be instances of particular roles. 2. 1 Reasoning Services DLs provide a variety of services [2] that make them particularly attractive as models for describing semi structured and complex information [5]. Subsumption The power of DLs is derived from the automatic determination of subsumption between compositional descriptions. Given two conceptual definitions A and B, we can determine whether A subsumes B, in other words whether every instance of B is necessarily an instance of A. Formally, ....

[Article contains additional citation context not shown here]

A. Borgida. Description Logics in Data Management. IEEE Transactions on Knowledge and Data Engineering, 7(5):671--782, 1995.

....the frame target is classi ed in the hierarchy H idx . Every source case the index of which subsumes the frame target can be possibly reused for solving the target problem. Case retrieval in H idx can be likened to query processing in a concept hierarchy H based on a description logic [ Borgida,1995 ] the query is represented by a concept Q that is classi ed in H; the set of potential answers is composed of every individual i being an instance of a concept subsumed by Q, or being an instance of the most speci c subsumers of Q and verifying the constraints attached to Q. Retrieval of case ....

A. Borgida. Description Logics in Data Management. IEEE Transactions on Knowledge and Data Engineering, 7(5):671682, 1995.

....cooperative information system (CIS) described in [8] A CIS is quite similar to a federated databases. It is composed of a set of database schemata and a set of so called interschema assertions. Database schemata represent the individual information sources and are theories in description logics [6]. Interschema assertions formalize relations between different database schemata. CISs formalize a certain degree of autonomy, having each database its own language, domain, and schema. Furthermore CISs formalize a certain degree af redundancy by means of four different kinds of interschema ....

A. Borgida. Description Logics in Data Management. IEEE Transactions on Knowledge and Data Engineering, October 1995.

....implementable, specialised and automated deductive services 3 . However, although useful for dealing with class centred conceptual models such as the entity relationship model, existing DLs are unable to capture homogeneous schemas 4 based on behavioural abstractions such as processes [10, 11]. Moreover, they are unable to capture heterogeneous schemas 5 based on multiple intertwined abstractions considered at the same time (structures, behaviours, objects, etc. The EDDL (Extensible Domain Description Language) approach [12, 13, 14] has been recently introduced to allow the capture ....

A. Borgida. Description Logics in Data Management. IEEE Transactions on Knowledge and Data Engineering, 7(5):671682, 1995.

....what are the problems encountered This second question deals with the workshop topic Integration of relational, deductive, and object oriented formalisms . Analogous questions about integration involves the study of relations between dbms and description logics (dl) Borgida and Brachman,1993] [Borgida,1995] , relations between objectoriented formalisms and dl [Napoli et al. 1994] and the management of large knowledge bases [Karp and Paley,1995] This paper is organized as follows: first, we briefly introduce data mining in the field of medicine, then we present object based representation systems, ....

.... [Napoli,1995] 4 DISCUSSION AND CONCLUSION There are numerous advantages of using obrs for data mining, e.g. conceptual model of the domain studied, validation and organization of queries and views, query and result reification [Borgida and Brachman,1993] these advantages are also discussed in [Borgida,1995] and [Karp and Paley,1995] An obrs system is an intermediate system with respect to object oriented, frame based and description logic systems. It has the advantage of combining numerous interesting properties of the previous systems, especially in a data mining perspective. Moreover, obrs are ....

A. Borgida. Description Logics in Data Management. IEEE Transactions on Knowledge and Data Engineering, 7(5):671--682, 1995.

....or configured semi automatically from two generic instruments defined with the meta model: query classes and matrix based visualizations. Query classes are parameterized and possibly materialized views defined by necessary and sufficient membership constraints similar to description logics [2]. They defer the checking of consistency and completeness of the perspectives generated by parallel development to a moment definable by the teams themselves. Usually, a large number of query classes can be associated with a meta model and later applied in the negotiations between the modeling ....

A. Borgida. Description Logics in Data Management. IEEE Transactions on Knowledge and Data Engineering, 7(5):671 -- 682, 1995.

....are implemented by concept languages, the most known ones being CLASSIC [6] Kris [2] and Loom [14] Practical applications of these languages cover a great variety of domains. In particular, there are applications for database schema modelling and more general knowledge based management systems [4, 3], conceptual modelling of information sources in global information systems [13] and proposals of terminological models for Information Retrieval [15, 20, 16] Resource Discovery Mechanisms are used across the Internet to easily access network available collections of data. They embody languages ....

A. Borgida. Description logics in data management. IEEE Transactions on Knowledge and Data Engineering, 7(5):671--682, 1995.

....with Predicate Calculus. From an applied perspective, research has been done in the implementation of Concept Languages (CLs) among them CLASSIC [8] is a well known tractable language [9] and practical applications of these CLs. In particular, there are applications for semantic data modelling [10] and proposals of terminological models of IR [2,11] DLs see the world as a set of objects, here called individuals. Concepts denote subsets of individuals and roles denote binary relations between individuals. Concept languages provide term constructors for building variable free composite terms ....

Borgida, A. Description logics in data management. "IEEE Transactions on Knowledge and Data Engineering", 7(5), 1995, 671-682.

....from the extended language. This is indeed the case with DATALOG ; it is easily (polynomially) recognizable whether a query does or does not contain disjunctions or unstratified negation. For a discussion of similar modular approaches in the field of terminological databases, cf. Section 6 in [13]; for a related discussion of the expressiveness versus complexity question, see [30] Finally, we believe that the exact determination of the expressiveness and complexity of DATALOG ; at the second level of the Polynomial Hierarchy is important for two reasons. The expressiveness results ....

A. Borgida. Description Logics in Data Management. IEEE Transactions on Knowledge and Data Engineering, 7(5):671--682, 1995.

....cooperative information system (CIS) described in [8] A CIS is quite similar to a federated databases. It is composed of a set of database schemata and a set of so called interschema assertions. Database schemata represent the individual information sources and are theories in description logics [6]. Interschema assertions formalize relations between different database schemata. CISs formalize a certain degree of autonomy, each database having its own language, domain, and schema. Furthermore CISs formalize a certain degree of redundancy by means of interschema assertions, which capture four ....

A. Borgida. Description Logics in Data Management. IEEE Transactions on Knowledge and Data Engineering, October 1995.

....have been extensively studied (see e.g. 15] and [21] The following sections give a brief introduction to some aspects of DL theory. We do not attempt to give a thorough overview and formal account of DL theory. However, we try to summarize the notions important for this paper and refer to [5, 3, 11] for more complete information about description logic theory. 2.2.1 DL: The Abstract Domain In a DL a factual world consists of named individuals and their relationships that are asserted through binary relations. Hierarchical descriptions about sets of individuals form the terminological ....

A. Borgida. Description Logics in Data Management. IEEE Transactions on Knowledge and Data Engineering, 7(5):671--682, 1995.

....information system (CIS) described in [4] A CIS is quite similar to an information integration state. It is composed of a set of database schemata and a set of so called interschema assertions. Database schemata represent the individual information sources and are theories in description logics [3]. Interschema assertions formalize relations between different database schemata. CISs formalize a certain degree of autonomy, each database having its own language, domain, and schema. Furthermore CISs formalize a certain degree of redundancy by means of interschema assertions, which capture four ....

A. Borgida. Description Logics in Data Management. IEEE Transactions on Knowledge and Data Engineering, October 1995.

No context found.

A. Borgida. Description logics in data management. IEEE transactions on knowledge and data engineering, 7(1), 1995.

No context found.

A. Borgida. Description logics in data management. IEEE transactions on knowledge and data engineering, 7(1), 1995. 103

No context found.

A. Borgida. Description logics in data management. IEEE Transactions on Knowledge and Data Engineering. To appear.

No context found.

A. Borgida. Description logics in data management. IEEE Transactions on Knowledge and Data Engineering, 7(5):671--682, 1995.

No context found.

A. Borgida, Description Logics and Data Management, IEEE Transactions on Knowledge and Data engineering, 7(5), pp 671-782, 1995.

No context found.

Alex Borgida. Description logics for data management. IEEE Transactions on Knowledge and Data Engineering, 5(7), October 1995.

No context found.

A. Borgida. Description Logics in Data Management. IEEE Transactions on Knowledge and Data Engineering, October 1995.

No context found.

A. Borgida. Description logics in data management. IEEE Transactions on Knowledge and Data Engineering, 5(7):671--682, Oct. 1995.

No context found.

A. Borgida. Description logics for data management. IEEE Transactions on Knowledge and Data Engineering, 5(7), Oct. 1995.

No context found.

A. Borgida. Description Logics in Data Management. IEEE Transactions on Knowledge and Data Engineering, 7#5#:671#682, October 1995.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC