R. Hull, R. King. Semantic Database Modeling: Survey, Applications, and Research Issues. ACM Computing Surveys, 19(3):201--260, 1987.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....operations. The unnecessary and phantom preconditions can be omitted. Integrity constraints can be also used to optimize user queries. The theory of dependencies is discussed in [AbV85, AtD93, Mai83, PDG89, Tha91, Var88] The extension of this theory to other database model is developed in [AlT92, HuK87], Hul89, Jac82, Tha92] for semantic models and in [AbK91, BiD91, GKS91, Heu89, KiL89 ] STW91] for object oriented models. The book [Zal89] gives an impression on the algebraic treatment of dependencies. A survey on deductive databases is presented in [CGT91] Ull89] Further, the papers [Bid91, ....

R. Hull and R. King, Semantic database modeling: Survey, applications, and research isues. ACM Computing Surveys 19, 3, 1987, 201-260.

....codomains. It is possible to create sets of elements in ADAMS, but not sets of data values. Every ADAMS element must belong to a class. The class system supports multiple inheritance [Car84] In this regard, and in its syntax and usage, ADAMS is a semantic database system in the sense of [HuK87]. Probably the most important aspect of ADAMS is its treatment of names. Although there are many different ways of referencing desired data elements and their values [KhC86] at some fundamental level data access depends on the ability to name elements, or sets of elements, in the database. In ....

R. Hull and R. King, Semantic Database Modeling: Survey, Applications, and Research Issues, Computing Surveys 19,3 (Sep. 1987), 201-260.

....or discovered knowledge about databases, queries, and users. Therefore, they interact closely with the major components of the KRDB. 5. 1 Analysis of the intent of a query Several interesting methods for query intent analysis have been developed in the studies of intelligent query answering [14, 15, 19, 7, 26]. Such analyses are based on the notions of generalization, association, aggregation, concept clustering, etc. Semantic data modeling, classification of topics of interests, and plan analysis and formation are powerful techniques for query intent analysis [15, 19, 7, 26] When posing a query, ....

R. Null and R. King. Semantic database modeling: Survey, applications, and research issues. ACM Cornput. Surv., 19:201 260, 1987.

....the evaluation of modeling methodologies for information systems development is a very active area of research. Researchers within this area have recognized the need for systematic evaluations of modeling methodologies. A substantial number of evaluations are reported in the literature [1, 13, 18, 6, 10, 17, 22, 23, 24, 14, 25], and IFIP Working Group 8.1 is co sponsoring an annual workshop, EMMSAD, devoted solely to this topic. The outcomes of the evaluations are useful for both designers of new methodologies and the users of the methodologies: The designers can use the evaluation criteria as design criteria when ....

....evaluated and compared. Schrefl et al. 18] develop a set of criteria for comparing conceptual, or semantic, data models and evaluate seven conceptual data models with respect to these criteria. The criteria focus on which modeling constructs, a semantic data model should offer. Hull and King [10] discuss issues of conceptual data modeling and survey sixteen conceptual data models. The models are subdivided in five categories: prominent models (the most well known at the time) other highly structured models, binary models, and relational extensions. The focus of the survey is on the ....

R Hull and R. King. Semantic Database Modeling: Survey, Applications, and Research Issues. ACM Computing Surveys, 19(3):201--260, September 1987.

....historically, such as the Semantic Database Model (SDM) The term system used here is deliberately vague and is meant to include methodologies, implementations, standards, and specifications. 62 published by Hammer and McLeod in 1981, which emphasized the concept of derived schema components [78]. However, these earlier modeling languages only set the stage for later modeling developments. It was these later developments such as NIAM, IDEF0, IDEF1X, and EXPRESS which are more key to STEP development. The following section provides more background and functional description on each of ....

Hull, Richard, and King, Roger. Semantic Database Modeling: Survey, Applications, and Research Issues. ACM Computing Surveys, Vol. 19, No.3; 1987 September.

.... Entity Relationship Diagrams with Complex Attribute Dependencies Carsten Lutz LuFG Theoretical Computer Science, RWTH Aachen, Germany Email: lutz informatik.rwth aachen.de 1 Motivation Entity Relationship (ER) diagrams are among the most popular formalisms for the support of database design [6, 10, 5]. During the last years, the initially rather simple ER formalism has been extended by various means of expressivity to account for new, more complex application areas such as schema integration for data warehouses [10, 11] Designing a conceptual model with such enriched ER diagrams is a ....

.... are among the most popular formalisms for the support of database design [6, 10, 5] During the last years, the initially rather simple ER formalism has been extended by various means of expressivity to account for new, more complex application areas such as schema integration for data warehouses [10, 11]. Designing a conceptual model with such enriched ER diagrams is a nontrivial task: there exist complex interactions between the various means of expressivity, which quite often results in unnoticed inconsistencies in the ER schemas and in implicit rami cations of the modeling that have not been ....

R. Hull and R. King. Semantic database modeling: Survey, applications, and research issues. ACM Computing Surveys, 19(3):201-260, 1987.

.... conceptual model, the Spatio Temporal Unifying Semantic Model (ST USM) Geographic information is defined as having a theme (i.e. the phenomenon or object being observed) the location of the phenomenon and the time related to the phenomenon [72] Conventional conceptual models (e.g. [11, 16, 31, 56]) do not provide a mechanism to explicitly capture the semantics related to space and time; as a result, the users spatio temporal data requirements cannot be adequately captured during conceptual design. Consequently, many researchers [10, 20, 22, 23, 42, 46] cite the need for abstract concepts ....

....supports spatial abstractions. We round out the paper with a comparison with extant spatio temporal conceptual models and describe our future research directions. 4 2 Motivation The representation of geographic phenomena in databases is one of the central issues in GIS [45] Conceptual modeling [11, 16, 31, 56] takes the user requirements as an input and transforms them into a high level conceptual schema. A good conceptual schema acts as glue between users, data analysts and the database implementation. While designing a database, a data analyst may use the conceptual schema as a communication tool ....

R. Hull and R. King, Semantic Database Modeling: Survey, Applications, and Research Issues,ACM Computing Surveys, 210-260, 1987.

....tree model has been used. 8] provides also an extension of tree inclusion problems by logical variables used to extract substructures of the pattern instances and to express equality constraints on them. The work presented here is also related to semantic modeling in general, see e.g. [12] and [14] and especially to ontologies and RDF(see Section 3) ....

R. K. Richard Hull. Semantic database modeling: Survey, applications, and research issues. ACM Computing Surveys, 19(3):201--260, September 1987.

.... technique is the entity relationship (ER) modeling, which was introduced by Peter Chen in 1976 and has been extended many times since then (Date 1986; Bowers 1988) The most recent extensions to the ER model were presented in 1986 by Batini, Lenzerini, and Navathe and by Teorey, Yang and Fry (Hull and King 1987) and in 1989 by Elmasri and Navathe. The ER model includes the concepts of entity set, relationship (Fig. 1) and attributes (Bowers 1988; Batini et al. 1992) In this study the basic ER model was used in conceptual analyses and the relational model for the implementation of the database. Five ....

Hull, R., and King, R. 1987, Semantic database modeling: survey, applications and research issues. ACM Computing Surveys, 19(3): 201--260. ACM press, New York.

....etc) models. Semantic models allow a natural and intuitive description of the real world because they allow the end users including database designers to think of data in a way which relates more closely to real world data. A comprehensive survey of the area of semantic modeling may be found in [HUL87] and [PEC88] HUL87] contains a general semantic model referred to as GSM intended to be representative of a class of semantic models. The popular semantic models include the E R model, the functional data model (FDM) and the semantic database model. Further details on semantic data models may ....

....models allow a natural and intuitive description of the real world because they allow the end users including database designers to think of data in a way which relates more closely to real world data. A comprehensive survey of the area of semantic modeling may be found in [HUL87] and [PEC88] HUL87] contains a general semantic model referred to as GSM intended to be representative of a class of semantic models. The popular semantic models include the E R model, the functional data model (FDM) and the semantic database model. Further details on semantic data models may be found in [HAM81, ....

R. Hull and R. King, Semantic Database Modeling: Survey, Applications, and Research Issues, ACM Computing Surveys, Vol 19, Number 3, September 1987.

....then adapt the optimized form generated at the metalevel to this particular instance. Our results, which concentrate on the aggregation and classification abstractions, complement the work on type inferencing (i.e. generalization hierachies) in object oriented database languages [KLW90,ABIT90,HK87b] and KL One like knowledge representation languages [BBMR89] it is our final goal to integrate these aspects in the same framework as presented here. 2. Relational Simplification Techniques Simplification techniques for integrity constraints within relational databases have first been proposed ....

Hull,R., King,R. (1987). Semantic database modeling: survey, applications, and research issues. ACM Computing Surveys 19(3), Sept. 1987.

....nition of an individual object and to derive the corresponding concept. The latter task which is some kind of type inference is sometimes treated separately by an inference mechanism called the recognizer. Type generalization specialization hierarchies are also used in semantic data models (SDMs) [7]. A prominent representative of SDMs is the IFO model [1] due to its genericity in the sense that almost all known SDMs are representable in IFO. A schema in IFO consists of a hierarchy of classes, where specialization by introducing subclasses di ers from generalization by union construction, and ....

....Boats : fMotor Boatg) 2.4 Semantics of IFO Schemata So far we have only de ned the syntax of an IFO schema. We have seen that it smoothly extends both IFO and TSLs. We want to extend our approach giving a model theoretic semantics for IFO extending the one for TSLs [5] and for IFO [1, 2, 7]. The general TSL approach starts with an arbitrary countably in nite set D. Each type gets mapped to a subset of D and each role gets mapped to a subset of D D. The mapping is usually denoted in both cases. Hence a type corresponds to an immutable set of possible values. Moreover, we have ....

R. Hull and R. King. Semantic database modeling: Survey, applications and research issues. ACM Computing Surveys, 19(3), September 1987.

....of an ooD approach. The data structure provided by the relational model is too simplistic for many applications. Although there is no accepted standard data model for ooD systems, these systems are in general based on richer structures previously studied in the so called semantic data models [30] and the object oriented approach brings improvements in data modelling. First, the notion of object elegantly fits data from the disk to the screen. For instance, in the O 2 database system, a persistent object can be displayed on the screen by a simple method application. This uniformity of ....

R. Hull and R. King. Semantic database modeling: Survey, applications, and research issues. ACM Computing Surveys, 19:201--260, 1987.

.... Object Oriented (OO) Database systems distinguish two notions on the model: the schema and the instances of the schema [18] Another (orthogonal) distinction concerns the structural and the behavioral part of an OO Database System [2] The structural part is quite similar to semantic modeling [17, 1]. The behavioral part deals with methods [18] The main components of the schema are the class hierarchy and the method signatures. This paper does not deal with behavioral aspects or instances. Rather, we combine the OO paradigm with the Deductive one in a study of structural properties of OO ....

R. Hull and R. King. Semantic Database Modeling: Survey, Applications and Research issues. ACM Computing Surveys, 19(3):201--260, 1987.

....sugared COOL syntax for illustration only. BCOOL consists of only the few following concepts, which can also be found in (almost) all object oriented models: Types can be basic or constructed. Basic types describe (pairwise) disjoint sets of instances. Besides concrete (or printable [11]) types describing data, or values, such as integers, boolean, strings) there is a basic type denoting objects, on which object types can be defined by subtyping (see below) Objects are fully encapsulated. They can be used and manipulated only by means of 2 their interface, a set of functions. ....

R. Hull and R. King. Semantic database modeling: Survey, applications, and research issues. ACM Computing Surveys, 19(3):201--260, September 1987.

....constructor was avoided in [5] for safety reasons, but is certainly useful for many applications. Also, we consider methods . The various features that can be found in the data structure are standard in database models. See, for instance, the survey on Semantic Database Models of Hull and King [25] or recent proposals on object oriented database models, e.g. 12, 22, 33, 34] 2 The language that we propose is rule based. An important aspect is that it is (to a large extent) declarative . All the user can do is specify statements of the forms: if condition then assertion ; ....

R. Hull and R. King. Semantic database modeling: survey, applications, and research issues. ACM Computing Surveys, Sept. 1987.

.... Database Systems as Picture Retrieval Systems There has been a great interest in providing several extensions to the relational data model to overcome the limitations imposed by the at tabular structure of relations for geometric modeling, engineering and geographic applications [102, 86, 152, 80, 149, 103, 23]. The resulting data model is characterized by the addition of application specic components to an existing database system 11 Figure 2: System Architecture for Image Processing Graphics Systems with Database Functionality as Picture Retrieval Systems 12 Query Interface for Attribute Based ....

R. Hull and R. King. Semantic database modeling: survey, applications, and research issues. ACM Computing Surveys, 19(3):201-260, 1987.

....models. 2 2 The Extended Entity Relationship Model In this Section, we sketch briefly the main concepts of the Extended Entity Relationship (EER) model [15] 2 . It is based upon the classical Entity Relationship model [9] and extended with the following concepts known from semantic data models [24]: ffl components, i.e. object valued attributes to model complex structured entity types: ffl multivalued attributes and components to model association types; ffl the concept of type construction in order to support specialization and generalization; ffl several structural restrictions like ....

R. Hull and R. King. Semantic database modeling: Survey, applications, and research issues. ACM Comput. Surv., 19(3):201--260, 1987.

....responsible for these tests and therefore different application programs test individually. In this case contradictory tests are possible. There is intensive research (and commercial products are already on the market) to enrich the capabilities of modern systems to enforce stricter tests [14] [8], 11] 7] 15] But even in a modern IS, capable to perform sophisticated tests, performance reasons often prevent their application (e.g. it is possible to detect similar entities by enforcing a unique index computing an edit distance on an appropriate data field [17] Normally the person ....

Richard Hull and Roger King. Semantic Database Modeling: Surveys, Applications, and Research Issues. ACM Computing Surveys, Vol. 19, No. 3, September 1987, pages 201--260, 1987.

....equivalence that hold between two data structures expressed in that model. 4.1 The Data Representation Model The data representation model encompasses atomic data types (like integer, AEoat, string, etc) and constructed types. We consider three type constructors commonly admitted in semantic [16] and object oriented data models [3] i) Labelled tuple constructor, denoted as X(Role 1 : Type 1 , Role n : Type n ) the tuple constructor is similar to the record type in programming languages like Pascal. Further, X (a 1 : r 1 ; Delta Delta Delta ; a n : r n ; Delta Delta ....

R. Hull and R. King. Semantic Database Modeling : Survey, Applications and Research Issues. ACM Computing Surveys, 19(3), September 1987.

....generalized join and projection in its polymorphic type system. Sets may be constructed on any description type. Combined with labeled records, labeled variants and cyclic definitions, the Machiavelli type system allows us to represent most of the structures found in various complex data models [HK87] Cyclic structures are supported by exploiting the properties of regular trees [Cou83] Join and projection are generalized to arbitrary, possibly cyclic, structures and are polymorphic functions in Machiavelli s type system. Complex object or non firstnormal form relations are usually ....

R. Hull and R. King. Semantic database modeling: Survey, applications and research issues. Computing Surveys, 19(3), September 1987.

....the best complexity fit. With the tremendous growth of data stored in large databases, it becomes imperative to develop efficient methods to perform classification on such huge amounts of data. Although data in a relational databases are usually well formatted and modeled by semantic data models [43], the contents of the data may not be classified. For instance, a biological database may store a large amount of experimental data in a relational format, but it is still necessary to classify the data in order to determine the intrinsic regularity of the data. Since schemas and data formats ....

R. Hull and R. King. Semantic database modeling: Survey, applications, and research issues. ACM Comput. Surv., 19:201--260, 1987.

....only semantically correct modifications are made to the database; and the inability to define a new entity type as a specialization of an existing entity type. From the mid eighties, various extensions to relational databases and alternative types of systems have been proposed and implemented [13, 55, 83, 95]. Each of these alternatives had different goals, but most attempted to provide a richer data model and address the impedance mismatch problem by providing a uniform language for representing data, views, constraints, queries, and application programs. Two of these extensions are deductive ....

R. Hull and R. King. Semantic database modeling: Survey, applications and research issues. ACM Computing Surveys, 19(3):201--260, 1987.

....activitites in the database community, PROGRES specifications can be viewed as conceptual database model specifications. They determine the static structure as well as the allowed dynamic behavior of a data base. Our underlying data model has a lot of similarities to so called semantic data models [28], and especially to the binary relationship model [15] However, while in the database area the specification is usually restricted to modeling the static structure of a database or to define queries for an already existing database [2,1] our specification language also offers sophisticated ....

R. Hull and R. King. Semantic database modeling: Survey, applications, and research issues. ACM Computing Surveys, 19(3):201--260, 1987.

....activitites in the database community, PROGRES specifications can be viewed as conceptual database model specifications. They determine the static structure as well as the allowed dynamic behavior of a data base. Our underlying data model has a lot of similarities to so called semantic data models [28], and especially to the binary relationship model [29] However, while in the database area the specification is usually restricted to modeling the static structure of a database or to define queries for an already existing database [30,3] our specification language also offers sophisticated ....

R. Hull and R. King. Semantic database modeling: Survey, applications, and research issues. ACM Computing Surveys, 19(3):201--260, 1987.

....they lack some of the desired properties for an object oriented model. Since the object oriented paradigm already possess some of the useful features essential for next generation data 1 See [63] for a detailed survey of SDMs. A detailed reading and a tutorial on SDMs may also be found in [39]. 6 and knowledge base applications, it is believed that semantically rich and expressive conceptual models can be built for future applications by extending the functionality of the SDMs in the direction of object oriented programming. The resulting model would likely become similar to abstract ....

....and obviously other important features. The DERDL model [37] also captures a few features in our direction. Particularly, it proposed the idea of dynamic objects and relationships, and similar other dynamic properties of database that is close to the spirit of derived schema components in SDM [36, 63, 39]. Since the underlying concept is value based, it also has the shortcomings of the DK model. The OBJECTModeler presented in [27] attempts to capture user s view of the data and takes an OO approach. Similar to ER model, it emphasizes on type constructors instead of attributes and functions. It ....

[Article contains additional citation context not shown here]

R. Hull and R. King. Semantic database modeling: Survey, applications, and research issues. 19(3):201--260, September 1987.

....a software system, such as the design of object structures, the effect of operations on objects, or the synchronization of concurrently executed tasks. Many of these languages use special classes of graphs as their underlying data models. Conceptual graphs [Sowa 84] semantic) data base models [HK 87] petri nets [GJRT 82] or attributed trees [Reps 84] are well known examples of this kind. 1. Supported by Deutsche Forschungsgemeinschaft. 2. Supported by Stiftung Volkswagenwerk and the Bundesministerium f r Forschung und Technologie. 3. cf. Nagl 80] EJS 88] Lew 88] Appears in: ....

Hull R., King R.: Semantic Database Modeling: Survey, Applications, and Research Issues; in: ACM Computing Surveys, vol. 19, No. 3, acm Press (1987), pp. 201-260

....correctly notes that inheritance as defined by the IS A construct is really little more than a convenient syntactic shorthand for incrementally creating subclasses. We, too, will treat it in just this fashion. This way of incrementally creating subclasses has also been called specialization [HuK87]. number, and can therefore serve as an operand wherever a real is expected. This inheritance property is not reflected in the bit string representations of these types. One may not use an integer bit string where a real bit string is expected. The integer must first be explicitly coerced into a ....

.... social security nbr , and address ; the subclass hierarchy of Figure 2 2 would become C PATIENT DOCTOR PERSON Generalized subclass hierarchy Figure 2 5 and class of the union would be S [C ] The creation of such most general upper bounds in the subclass hierarchy has been called generalization [HuK87]. While it leads to a consistent semantic interpretation, it appears to be fairly difficult to implement in practice. The intersection operation presents different problems. Suppose we wish to process the set of residents who have themselves been trauma patients , say in a loop of the form ....

R. Hull and R. King, Semantic Database Modeling: Survey, Applications, and Research Issues, Computing Surveys 19,3 (Sep. 1987), 201-260.

.... we would rather say Complex Values) 3) finally, several proposals have been made to include other type constructors (such as lists, multisets, arrays, Figure 1 gives a graphical summary of (1NF) relations, nested relations, and Complex Objects using the notation from [56]. a) b) c) Figure 1: Flat Relations (a) Nested Relations (b) and Complex Objects (c) Example 1 The following is an example of three schema definitions in an imaginary syntax using record and set as constructors. In (a) we see the traditional flat relations ....

.... of type definitions that was recursive (c) It was not a priori clear, what such recursive type definitions should mean, and the nested relational approaches that use equations for the definition of relational schemas explicitly rule out recursion [44, 93] Also, the generic semantic data model of [56] does not permit recursive use of the set and or tuple constructors (association and aggregation) the construction shown in Figure 2(a) is not allowed. The situation has to be modeled using at least one function between types. In Figure 2(b) we have shown a valid modeling using two functions ....

R. Hull and R. King. Semantic database modeling: Survey, applications, and research issues. ACM Computing Surveys, 19(3):201--260, September 1987.

....objects is not explicit for users, handles are needed in order to refer to objects (e.g. newly created ones) Therefore we integrate variables. We now define the type system and type valid (query )expressions and update operations. Type Expressions. Types can be atomic or constructed (see [HK87] The set of atomic types ( contains predefined (or printable) ones such as INTEGER, BOOL (denoted by Int ; Bool ) and one type that represents the countable set of objects Object . Object types need not be named, they are denoted by function labels ( f 1 ; f n ] Type constructors ....

....ff 00 1 ; f 00 l g = ff 1 ; f n g [ ff 0 1 ; f 0 m g: 3 Because all active domains are finite sets, the subset relationship is always proper. 4 In case that objects are not allowed to become instances of arbitrary other types as in the case with abstract types in [HK87] or clusters in OSCAR [HFW90] our approach can be extended to different sub lattices of objects. This would require more complex semantics of set operations, since both input sets have to belong to the same cluster (see also Sec. 5.3) 5 Notice that this definition captures inheritance of ....

[Article contains additional citation context not shown here]

R. Hull and R. King. Semantic database modeling: Survey, applications, and research issues. ACM Computing Surveys, 19(3):201--260, September 1987.

....definable codomains. It is possible to create sets of elements in ADAMS, but not sets of data values. Every ADAMS element must belong to a class. The class system supports multiple inheritance [Car84] In this regard, and in its syntax and usage, ADAMS is a semantic database system in the sense of [HuK87]. Probably the most important aspect of ADAMS is its treatment of names. Although there are many different ways of referencing desired data elements and their values [KhC86] at some fundamental level data access depends on the ability to name elements, or sets of elements, in the database. In the ....

R. Hull and R. King, Semantic Database Modeling: Survey, Applications, and Research Issues, Computing Surveys 19,3 (Sep. 1987), 201-260.

....nick cs.uregina.ca. Both database contents and database structures (schemes) may evolve over the lifetime of a database. Issues on the evolution of database structures (called schema evolution) have been studied in multidatabase, heterogeneous database and object oriented database research [1, 2, 3]. To avoid introducing an extra dimension of complexity, we focus our study on the evolution of database contents and assume that the database schemes are stable, not evolving over time. Moreover, we assume that the databases to be studied are relational ones. The extensions of our technique to ....

....France small 30 Workstation Canada moderate 10 Graphics U.S. big 15 PC U.S. small 35 Graphics Canada moderate 20 troubled firm Network Japan big 60 PC U.S. moderate 20 Main frame U.S. big 15 Table 6: A generalized relation for computer companies. The range for d weight is in the interval [0, 1]. A high d weight indicates that the concept is primarily derived from the target class C j , and a low d weight implies that the concept is primarily derived from the contrasting class(es) The process for learning discriminant rules is illustrated in Example 5. Example 5 Suppose a discriminant ....

R. Hull and R. King. Semantic database modeling: Survey, applications, and research issues. ACM Comput. Surv., 19:201--260, 1987.

....seen by the application and the user. To meet new requirements there was developed a diversity of languages and techniques for data definition and manipulation ranging from nested relational to object oriented semantic models; the latter are based on comprehensible graphical schemas (see surveys [25, 24]) However, it is commonly recognized (M.Atkinson et al.[2] C.Beery [6] S.Navathe [35] etc) that they lack formal semantics and, from an overall perspective, a peculiarity of the situation consists in extremely strong experimental activity based on surprisingly weak formal semantic foundations. ....

R. Hull and R. King. Semantic database modeling: Survey, applications and research issues. ACM Computing Surveys, 19(3):201--260, 1987.

....where we would not expect to be able to define equivalent (terminating) functions in more general programming systems. Another purpose of this work is to study the semantics of data models based on objectidentity. Object identities occur naturally in object oriented and semantic data models ([7, 5, 21]) and have been shown to enhance the expressive power of purely value based models and languages ( 2, 20] They provide an abstract model for our intuitions about how complex structures, and particularly cyclic structures, are represented in a database. There are a number of other equivalent ....

....will also support a number of other different kinds of constraints as primitive in the data model. For example relational databases will often support keys and sometimes functional and inclusion dependencies ( 36] while semantic models might incorporate various kinds of cardinality constraints([21]) In general such collections of constraints form a rather ad hoc selection, included because of their utility in the particular examples of databases that the designer of the system had in mind, rather than because of any theoretical justification. With the increasing complexity of data models, ....

Richard Hull and Roger King. Semantic database modeling: Survey, applications, and research issues. ACM Computing Surveys, 19(3):201--260, September 1987.

....Figure 3: Fragment of a domain model for financial reporting. Figure 3 shows a fragment of a domain model for an integration scenario in which financial reporting data is being exchanged across disparate systems. Each node in the graph represents an object. In the tradition of semantic data models [9], we distinguish between abstraction objects representing things in the real world and those which are descriptive features or attributes of those things; we refer to these as entity types (of which instances are simply entities) and semantic domains (respectively, semantic values) ....

Hull, R., and King, R. Semantic database modeling: survey, applications, and research issues. ACM Computing Surveys 19, 3 (1987), 201--260.

....and role specialization semantic hierarchies. 3 Exploiting Specialization Generalization Relationships for User Assistance In the area of Semantic Data Modeling semantic hierarchies have long ago been borrowed from the area of Artificial Intelligence and adapted to the needs of databases [6]. In this section we briefly recall a number of basic features that have been identified with respect to specialization generalization hierarchies and outline how these features are exploited for user assistance by query modification. 3.1 Basic constraints A conventional way to model the ....

....of objects, to which base types are attached. If specialized classes are desired, they are derived from other classes by means of e.g. the ISA relationship. For two classes in a ISA relationship, their extensions are in a subset relationship and their types are in a subtype relationship. In [6] constraints are identified that ensure consistency of ISA relationships on classes, in particular it follows that they are hierarchical structures. In addition to the basic constraints, other constraints may be imposed optionally, thus characterizing different kinds of ISA hierarchies, e.g. 1) ....

Hull, R., King, R.: Semantic Database Modeling: Survey, Applications and Research Issues. ACM Computing Surveys, 19, 3, 201--260 (1987).

....the need for EDBs and present them using some simple examples demonstrating the translation to EA rules. Our goal is to faithfully capture richer data semantics in a constantly changing environment. Object oriented [4, 10] active [8, 26] deductive [13] functional [32] semantic data models [19, 27], graph based [9, 3, 24] temporal [7, 14, 22, 28] spatial [23, 11, 28] and many other database technologies all have as their primary focus the layering of more human semantics over the traditional symbolic representations of data. The effects of these efforts, in terms of expressive power, ....

R. Hull, R. King, Semantic Database Modeling : Surveys, Applications and Research Issues, ACM Computing Surveys, Vol 19, Sept 1987.

....and include: 1. The AND approach. The AND approach views all the values as applicable, thus a retrieval request would result in the set fv 1 ; v n g. This approach was referred to in semantic data models as a multi valued attribute, in which a data item s value consists of several values [HR87] For example, a data item that designates the languages that a person speaks, can have a set of grouped values. The interpretation is AND, designating that the person speaks all the languages in the designated set. 2. The possible world (XOR) Approach. Under this approach, there exists a ....

R. Hull and R.King. Semantic database modeling: Survey, application and research issues. ACM Computing Surveys, 19(3):201--260, Sep 1987.

No context found.

R. Hull and R. King. Semantic database modeling: Survey, applications, and research issues. ACM Computing Surveys, 19(3):201--260, September 1987.

....LAR execution model explores a directed acyclic graph (DAG) of possibilities, where a fork in the DAG corresponds to the firing of a rule whose action involves a disjunction. In [CHM96] LARs and their associated execution model are applied to updating derived data in semantic models (see, e.g. HK87] for a description of semantic data models) As a simple example, suppose that the class Employed Student is defined to be the intersection of classes Employee and Student, that x is in the class Employed Student, and that an explicit request to delete x from Employed Student has been made. This ....

R. Hull and R. King. Semantic database modeling: Survey, applications, and research issues. ACM Computing Surveys, 19(3):201--260, September 1987.

No context found.

R. Hull, R. King. Semantic Database Modeling: Survey, Applications, and Research Issues. ACM Computing Surveys, 19(3):201--260, 1987.

No context found.

Hull R., King R.: Semantic Database Modeling: Survey, Applications, and Research Issues; in: ACM Computing Surveys, vol. 19, No. 3, acm Press (1987), pp. 201-260

No context found.

R. Hull and R. King. Semantic Database Modeling: Survey, Applications, and Research Issues. ACM Computing Surveys, 19(3):201--260, 1987.

No context found.

Richard Hull and Roger King. Semantic Database Modeling: Survey, Application, and Research Issues. ACM Computing Surveys, 19(3), September 1987.

No context found.

R. Hull and R. King. Semantic database modeling: Survey, applications, and research issues. ACM Computing Surveys, 19(3):201-260, 1987.

No context found.

Richard Hull and Roger King. Semantic database modeling: Survey, applications, and research issues. ACM Computing Surveys, 19(3):201--260, 1987.

No context found.

Hull, R. & King, R. (1987). Semantic database modeling: survey, applications, and research issues. ACM Computing Surveys, 19(3), 201-260.

No context found.

R. Hull and R. King. Semantic database modeling: Survey, applications and research issues. ACM Computing Surveys, 19(3):201--260, 1987.

No context found.

R. Hull and R. King, #Semantic Database Modeling: Survey, Applications, and Research Issues," ACM Computing Surveys, Vol. 19, No. 3, pp. 201#260, 1987.

No context found.

R. Hull and R. King. Semantic database modeling: Survey, applications, and resource issues. ACM Computing Surveys, 19(3):201--260, September 1987.

First 50 documents  Next 50

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