C.M.R. Leung and G.M. Nijssen. Relational database design using the NIAM conceptual schema. Information Systems, 13(2):219--227, 1988.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....a prototype object base in which the transactions and queries mentioned in section 3 above can be executed. The other kind of application consists of testing CMSL on examples. This includes taking examples from diverse system analysis and development methods, like JSD [19] ISAC [22] NIAM [21, 24], and Structured Analysis [9] and translating them into CMSL. This gives us a clue about the kinds of tasks and heuristics mentioned in equation (2) and will bring us closer to achieving our aim of finding a solution to these equations. Finally, on the theoretical side we plan to start work on ....

C.M.R. Leung and G.M. Nijssen, "Relational Database Design Using the NIAM Conceptual Schema," Information Systems 12(2), pp. 219-227 (1988).

....schemas depicted are mathematically equivalent (for a proof of this, refer to [Hal89] These two alternatives however might not be equally preferable ways for the user(s) to think about the application. The schemas in the example are modelled using the Object Role Modelling (ORM) technique ( LN88] HW93] Win90] Hal95] Similar examples for EER or OO models could be given. Once the user has selected the preferred conceptual schema, this should be used in any later conceptual queries on the implemented information system, assuming the availablity of an appropriate conceptual query ....

C.M.R. Leung and G.M. Nijssen. Relational database design using the NIAM conceptual schema. Information Systems, 13(2):219--227, 1988.

....have direct pendants in the used information modelling technique. This does not only warrant a natural way of modelling, but also maintains the rich semantics of verbalisations in natural language as much as possible. Among the first to apply this principle to conceptual modelling was Nijssen [48, 54, 44]. The information modelling technique we use throughout this paper is a modernised version of NIAM, called Object Role Modelling (ORM) Before continuing we give a brief account of the history of this ORM version. In [5] the Predicator Model was introduced as one of the first formalisations of ....

C.M.R. Leung and G.M. Nijssen. Relational database design using the NIAM conceptual schema. Information Systems, 13(2):219--227 (1988).

....domain by the development team and users. A conceptual model should describe both the static and the dynamic aspects of a piece of reality for which the information system is developed. In the last decade, many conceptual modeling approaches have been proposed (see, e.g. anto81a] falk90a] [leun88a] [meer88b] saka84a] However, most approaches model these two aspects separately: the static aspect is modeled in a snapshot model, while the dynamic aspect is modeled in a process model [kung83a] Only some approaches (e.g. jung91a] sern89a] support integrated modeling of the static and ....

C. Leung and G. Nijssen, "Relational database design using the NIAM Conceptual Schema," Information Systems, Vol. 13, No. 2, 1988.

....y This work has been partially supported by the ESPRIT project APPED (2499) realisation can be (automatically) derived, using a specification language that is more machine oriented. Several conceptual modelling techniques have an underlying object role structure (e.g. ER [5] and NIAM [11], 15] In these techniques a model consists of fact types, defined in terms of object types and roles. Depending on the power of the technique, fact types may be of any degree and can be treated as object types (objectified fact types) Furthermore, a distinction can be made between lexical and ....

....involving e.g. textual data, computer aided design or image processing. Current approaches to the transformation of conceptual models into internal models focus mainly on the relational model. The result of the transformation is a relational schema in a certain normal form (see for instance [11], 13] 17] 21] 22] and [24] Other approaches can be found in [2] 10] 20] and [25] In this paper we present a framework for the transformation of object role models. We describe a representation mechanism for internal structures, such that the conventional internal models can be ....

C.M.R. Leung and G.M. Nijssen. Relational database design using the NIAM conceptual schema. Information Systems, 13(2):219--227, 1988.

....several (different) databases for the same conceptual data model (multiprototyping) 1 Introduction Object role models are widely used for conceptual data modelling. Although several approaches exist for converting an object role model into a good internal representation (see e.g. 11] [13], 15] 17] 18] 8] and [9] there is no general strategy for examining the time space and update retrieval trade off for a given database application. However, if the actual implementation of a given conceptual model must satisfy the requirements of a particular application environment, such ....

....of the underlying objectrole structure may be found in [2] while a detailed treatment of the corresponding internal representations is found in [5] and [3] 2. 1 Basics We assume that the reader is familiar with data modelling techniques having an underlying object role structure (e.g. NIAM [13], 15] and BRM [12] 18] The basic notions are informally summarized as follows. An object role data model consists of an information structure and a set of constraints on the possible populations of the information structure ( 2] The central notion is the predicator, defined as a role ....

C.M.R. Leung and G.M. Nijssen. Relational database design using the NIAM conceptual schema. Information Systems, 13(2):219--227, 1988.

....expected usage of the database. As a consequence, these strategies ignore the integrity rules (constraints) specified in the conceptual data model. Profile based evolution strategies are thus unable to find internal representations in a specific normal form, such as BCNF (e.g. 29] and ONF (e.g. [17]) Although we are interested in finding efficient internal representations, rather than internal representations in a specific normal form, it is desirable allowing a database designer to combine profile based evolution strategies with evolution strategies converging to normal forms. Therefore ....

....of internal representations, the following distinction is made: Normalization algorithms (e.g. 10] 24] 28] These algorithms are initialized with a set of dependencies, e.g. functional dependencies. They result in a set of relation types in a specific normal form. Grouping algorithms (e.g. [17], 22] 9] The input of these algorithms is a conceptual data model consisting of fact types. These fact types are grouped together in relation types, thus producing a set of relation types in a specific normal form. Preparatory algorithms (e.g. 13] 12] 29] The input and output of these ....

[Article contains additional citation context not shown here]

C.M.R. Leung and G.M. Nijssen. Relational database design using the NIAM conceptual schema. Information Systems, 13(2):219--227, 1988.

....the relational model described in [VASS80] need two different algorithms; one for each direction. Similarly, the mappings from E R, EER, or E 2 R to the relational, hierarchical, or network 6 model [ELMA89,OZKA90] from Iris to the relational [LYNG87] from NIAM to the relational model or ACS [LEUN88, FLYN85], from the relational model to the network model or its reversal [LARS83] etc. all need specific translation algorithms. Since these direct translation algorithms are model specific, there is no uniform approach to support algorithm development for different data models. As a result, the ....

C. M. R. Leung and G. M. Nijssen, "Relational database design using the NIAM conceptual schema," Information systems, vol. 13, no. 2, 1988.

.... 1, 6525 ED Nijmegen, The Netherlands, E mail: pvb cs.kun.nl y This work has been partially supported by the ESPRIT project APPED (2499) In this paper we restrict ourselves to (the broad class of) conceptual modelling techniques with an underlying object role structure (e.g. ER [4] and NIAM [14], 17] The main contributions of this paper are: 1. Definition of candidate database structures in terms of the conceptual schema at hand, as a suitable representation for individuals. 2. Introduction of evolution operators in terms of this internal representation. We give a global description ....

C.M.R. Leung and G.M. Nijssen. Relational database design using the NIAM conceptual schema. Information Systems, 13(2):219--227, 1988.

....is provided to the internal level. The components are integrated by a central notion: the predicator. Further integration is obtained by means of basic population rules. 1 Introduction Database transformations have been extensively studied in literature (see e.g. 2] 3] 16] 20] [21], 22] It is found that these transformations involve far more problems than just schema translation. These problems are further increased by the fact that very often the source database model is a conceptual model, which is especially suited to express data semantics, while the target data ....

....system (cf. 15] As conceptual and internal models are used for different purposes, their capabilities are also quite different. We summarize the possibilities of both kinds of models. Conceptual models directly represent the semantics of the data to be handled by the system (e.g. 8] 12] [21], 24] For this purpose a large variety of constraints is used along with generalization and specialization of object types and objectification of fact types (relationship types) Furthermore, high level information languages may be used for data definition and manipulation (see e.g. 11] 17] ....

[Article contains additional citation context not shown here]

C.M.R. Leung and G.M. Nijssen. Relational database design using the NIAM conceptual schema. Information Systems, 13(2):219--227, 1988.

....model has been specified in a suitable language, a proper realisation can be (automatically) derived, using a specification language that is more implementation oriented. Several approaches to transforming a conceptual model into an implementation are found in the literature (see e.g. 16] and [12]) However, the main problem in database design, i.e. finding an appropriate implementation for a given conceptual model, has in fact not yet been solved. In order to illustrate, this section 2 first considers the question why Evolutionary Database Optimization is necessary and how it can be ....

C.M.R. Leung and G.M. Nijssen. Relational database design using the NIAM conceptual schema. Information Systems, 13(2):219--227, 1988.

....an appropriate implementation may be expressed in terms of structural transformations leading to performance improvement. This however is not fully recognized in current Computer Aided Software Engineering tools, since usually only a standard structure is generated (e.g. Optimal Normal Form [19] [16]) Therefore, we have designed and implemented a framework for conversion and transformation of database structures, which provides a flexible search mechanism through the solution space of all possible structures for a given database application. The entire framework is described in [4] The main ....

....a framework for conversion and transformation of database structures, which provides a flexible search mechanism through the solution space of all possible structures for a given database application. The entire framework is described in [4] The main difference with related work (e.g. 1] 13] [16], 19] 22] is that transformations may be probabilistically guided, for instance on the basis of a multiobjective fitness function, or on the basis of other structural properties. This enables us to systematically examine properties of the solution space at hand, such as the time space ....

[Article contains additional citation context not shown here]

C.M.R. Leung and G.M. Nijssen. Relational database design using the NIAM conceptual schema. Information Systems, 13(2):219--227, 1988.

....mappings such as transformation of populations and operations, are beyond the scope of this paper. These are treated in [6] 2.1 Conceptual data models In this section we discuss conceptual data models. We focus on data modelling techniques with an underlying object role structure (e.g. NIAM [17], 18] and the Binary Relationship Model [16] 23] A formal definition of these models is found in [5] The basic notions are summarized as follows. An object role data model consists of an information structure I and a set of constraints C on the possible populations of the information ....

....process. Another example involves the evolution of simple internal representations, where each fact type is taken separately, into more complex internal representations where fact types are joined together. In this way specific Normal Forms may be constructed stepwise (e.g. the Optimal Normal Form [17], 18] such that partly normalized representations are yielded as intermediate solutions. Finally, guidance conditions may also be expressed in terms of storage requirements and average response time of the internal representations, for instance in hill climbing strategies. This will be ....

C.M.R. Leung and G.M. Nijssen. Relational database design using the NIAM conceptual schema. Information Systems, 13(2):219--227, 1988.

....strategy is to find an internal model having certain desirable characteristics, i.e. being optimal in some sense. Although several approaches exist for converting a conceptual data model into a good internal representation (e.g. based on functional dependencies and other constraints such as in [17], 21] 23] there is no general strategy for examining the time space and update retrieval trade off for a given database application. However, if the actual implementation of a given conceptual model must satisfy the requirements of a particular application environment, such a strategy is ....

....efficient or not (Section 2.3) 2.1 Conceptual data modelling In this section we discuss a broadly accepted technique for making conceptual descriptions of complex data structures (as encountered in administrative environments) viz. the object role data modelling technique (see e.g. 4] 8] [17] and [21] We clarify this by an example (see Figure 1) This information structure describes the structure of information to be recorded about projects and persons being involved in those projects. More precisely the information structure contains the following. A Project has a budget of a ....

C.M.R. Leung and G.M. Nijssen. Relational database design using the NIAM conceptual schema. Information Systems, 13(2):219--227, 1988.

No context found.

C.M.R. Leung and G.M. Nijssen. Relational database design using the NIAM conceptual schema. Information Systems, 13(2):219--227, 1988.

No context found.

C.M.R. Leung and G.M. Nijssen. Relational database design using the NIAM conceptual schema. Information Systems, 13(2):219--227, 1988.

No context found.

C.M.R. Leung and G.M. Nijssen. Relational database design using the NIAM conceptual schema. Information Systems, 13(2):219--227, 1988.

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