In this paper, we carefully explore the assumptions behind using information capacity equivalence as a measure of correctness for judging transformed schemas in schema integration and translation methodologies. We present a classification of common integration and translation tasks based on their operational goals and derive from them the relative information capacity requirements of the original and transformed schemas. We show that for many tasks, information capacity equivalence of the schemas is not strictly required. Based on this, we present a new definition of correctness that reflects each undertaken task. We then examine existing methodologies and show how anomalies can arise when using those that do not meet the proposed correctness criteria. 1 Introduction Formal work on schema equivalence has largely been ignored within practical schema integration and translation tools. Practitioners have felt that theoretical work is too narrow in scope to be applicable to the problems ...

this report, we examine the technical challenges to integration, critique the available tools and resources, and compare the cost and advantages of various methodologies. We begin by analyzing the basic steps in strict and complete integration: 1) transformation of the various schemas to a common data model; 2) matching of semantically related schema objects; 3) schema integration; 4) transformation of data to the federated database on demand; and 5) matching of semantically equivalent data. Some progress has been made on generic problems such as (1) and (3) within the wider database community, but issues of semantics (steps (2) and (5)) have only been dealt with any degree of success by domain experts within the biological community. We then look at the solution space of integration strategies as defined by two axes, the "tightness" of federation and the "degree" of instantiation, discuss where various solutions fall on this plane, and examine their cost and advantages/disadvantages. Finally, we examine technical challenges that are not -3- July 12, 1995

Current theoretical work offers measures of schema equivalence based on the information capacity of schemas. This work is based on the existence of abstract functions satisfying various restrictions between the sets of all instances of two schemas. In considering schemas that arise in practice, however, it is not clear how to reason about the existence of such abstract functions. Further, these notions of equivalence tend to be too liberal in that schemas are often considered equivalent when a practitioner would consider them to be different. As a result, practical integration methodologies have not utilized this theoretical foundation and most of them have relied on ad-hoc approaches. We present results that seek to bridge this gap. First, we consider the problem of deciding information capacity equivalence and dominance of schemas that occur in practice, i.e., those that can express inheritance and simple integrity constraints. We show that this problem is undecidable. This undecidab...

Although Entity-Relationship (ER) modelling techniques are commonly used for information modelling, Object-Role Modelling (ORM) techniques are becoming increasingly popular, partly because they include detailed design procedures providing guidelines for the modeller. As with the ER approach, a number of different ORM techniques exist. In this paper, we propose an integration of two theoretically well founded ORM techniques: FORM and PSM. Our main focus is on a common terminological framework, and on the notion of subtyping. Subtyping has long been an important feature of semantic approaches to conceptual schema design. It is also the concept in which FORM and PSM differ the most in their formalization. The subtyping issue is discussed from three different viewpoints covering syntactical, identification, and population issues. Finally, a wider comparison of approaches to subtyping is made, which encompasses other ER-based and ORM-based information modelling techniques, and highlights how formal subtype definitions facilitate a comprehensive specification of subtype constraints.

The need to transform data between heterogeneous databases arises from a number of critical tasks in data management. These tasks are complicated by schema evolution in the underlying databases, and by the presence of non-standard database constraints. We describe a declarative language, WOL, for specifying such transformations, and its implementation in a system called Morphase. WOL is designed to allow transformations between the complex data structures which arise in object-oriented databases as well as in complex relational databases, and to allow for reasoning about the interactions between database transformations and constraints. integrating the US Cities-and-States and European-Citiesand-Countries databases shown in Figures 1 and 2. The graphical notation used here is inspired by [2]: the boxes represent classes which are finite sets of objects; the arrows represent attributes, or functions on classes. name str

This paper was presented at OOER'95: 14 th International Conference on Conceptual Modeling, Springer LNCS, vol. 1021, pp. 191-203, and is reproduced here by permission

TRANSFORMING DATABASES WITH RECURSIVE DATA STRUCTURES Anthony Kosky Advisors: Susan Davidson and Peter Buneman. This thesis examines the problems of performing structural transformations on databases involving complex data-structures and object-identities, and proposes an approach to specifying and implementing such transformations. We start by looking at various applications of such database transformations, and at some of the more significant work in these areas. In particular we will look at work on transformations in the area of database integration, which has been one of the major motivating areas for this work. We will also look at various notions of correctness that have been proposed for database transformations, and show that the utility of such notions is limited by the dependence of transformations on certain implicit database constraints. We draw attention to the limitations of existing work on transformations, and argue that there is a need for a more general formalism...

Research on schema integration leads to the identification of many different conflict types. Some of them received much attention and many papers proposed solutions for their resolution. However, literature usually focuses on traditional problems, whilst new kinds of schema discrepancies, due to the object orientation or the generalization concept, are not really treated. Moreover, most of the proposed methodologies and strategies only allow binary comparisons between items to be integrated. This paper discusses n-ary (also called one-many) conflicts, and particularly the three fragmentation conflict types. These conflict types need specific operators for schema comparison. We propose a simple unified language for easy specification of these conflicts, and give many different techniques to solve them. We emphasize the benefit of separating the declaration of the correspondences from the choice of resolution technique. Our discourse is illustrated with the entity-relati...

Abstract. Database transformations arise in many di erent settings including database integration, evolution of database systems, and implementing user views and data-entry tools. This paper surveys approaches that have beentaken to problems in these settings, assesses their strengths and weaknesses, and develops requirements on a formal model for specifying and implementing database transformations. We also consider the problem of insuring the correctness of database transformations. In particular, we demonstrate that the usefulness of correctness conditions such as information preservation is hindered by theinteractions of transformations and database constraints, and the limited expressive power of established database constraint languages. We conclude that more general notions of correctness are required, and that there is a need for a uniform formalism for expressing both database transformations and constraints, and reasoning about their interactions. Finally we introduce WOL, a declarative language for specifying and implementing database transformations and constraints. We brie y describe the WOL language and its semantics, and argue that it addresses many of the requirements on a formalism for dealing with general database transformations. 1

This article presents the idea of defining a kernel for object role modelling techniques, upon which different drawing styles can be based. We propose such a kernel (the ORM kernel) and define, as a case study, an ER and a NIAM drawing style on top of it. One of the prominent advantages of such a kernel is the possibility to build a CASE-tool supporting multiple methods. Such a CASE-tool would allow users with different methodological backgrounds to use it and view the modelled domains in terms of their favourite method. This is illustrated using a running example of a concrete domain in which we use the ORM kernel in combination with the NIAM and ER drawing style. 1 Introduction In the last decades, a plethora of modelling techniques for the design of information systems has been developed (see e.g. [Bub86]). In general, these modelling techniques provide only a crude and incomplete description of their syntax, and semantics ([HW92]). This situation has led to the Methodology...