In this paper, we present a strategy to reduce the processing time needed for selection operations with many attributes in standard database systems. These problems mostly occur in data mining, data analysis, information retrieval, and applications with high combinatorial complexity. In these systems, standard indexes do not gain a satisfying performance. Currently, this problem is tackled using more computing power or special solutions instead of standard databases. Our approach is to interpret the queries as high-dimensional point or range queries. Thus, we provide a “real” solution to answering complex queries rather than merely postponing the problems using technical methods. We show the benefit of multidimensional data structures. This benefit can be transferred to a lot of applications (e.g. Business Intelligence, Bill Of Materials Explosion, DNA Sequence Search), not only advanced applications of database systems like GIS, CAD, or multimedia. Finally, a very small sample data set is used to show the profit of our approach and we present possible integration points of the ICIx into standard relational database.

To eliminate semantic conflicts among the heterogeneous databases involved in a cooperation, a set of correspondence assertions for declaring their semantic relationships have to be constructed by DBAs or by users. Normally, four set relationships between object classes: equivalence, inclusion, intersection, and exclusion will be defined to provide knowledge about correspondences that exist among the local schemas. In this paper, we would like to introduce a new assertion, the so-called derivation assertion, to accommodate more heterogeneities, which can not be treated by the existing methodologies. As an example, consider two local object-oriented schemas: S/sub 1/ and S/sub 2/ and assume that S/sub 1/ contains two classes: parent and brother, and S/sub 2/ contains a class: uncle. A derivation assertion of the form: S/sub l/(parent, brother)/spl rarr/S/sub 2/(uncle) can specify their corresponding semantic relationship clearly, which can not be established otherwise. We claim that this kind of assertions is necessary for the following reason. Imagine a query concerning uncle, submitted to the integrated schema from S/sub 1/ and S/sub 2/. If the above assertion is not specified, the query evaluation will not take schema S/sub 1/ into account and thus the answers to the query can not be correctly computed in the sense of cooperations.

In this paper, we consider the query evaluation problem in relational multidatabases and develop a method for generating optimal plans for queries submitted to such a system. Two aspects will be discussed: join tree balance and node allocation. For the first problem, we extend the approach for balancing a join tree proposed by Du et al, so that more balanced join trees can be obtained, when, we present the concepts of dynamic time tables and constrained topological order to do the node allocation so that both the current load states of local database systems and load changes during the join operations can be handled. In this way, the deficiency of Evrendilek's method can be removed.

Summary form only given. As the first step of database integration, a participating local database schema should be transformed into an abstract one to remove data model conflicts. For our project, the object-oriented schema is chosen to represent the integrated information and a semi-automatic method is developed to transform relational schemas into OO schemas. The main idea behind this method is to define two kinds of logics: a relational database logic L/sub db/ and an object-oriented logic L/sub o/ to formalize the corresponding data models. L/sub db/ is used to model a database structure as well as its state, while L/sub o/ is utilized for an object-oriented database. Then, based on these formalisms, a set of Horn-clause-like rules is constructed to perform meta-level reasoning for translating a formalism into another.

We consider the query evaluation problem in multidatabases and develop an algorithm for generating optimal plans for queries submitted to an integrated schema. We try to extend the basic transformation step used in the method proposed by (Du et al., 1994) and construct a dynamic time table to support a query optimization process. In this way, not only the join tree balance, but also the node allocation can be achieved according to both communication costs and load measurements. In addition, the queuing analysis has been utilized to estimate the response times of local database systems.

Interoperability of programming languages with persistency requires new concepts to make a single knowledge base management system (KBMS), a custom-tailored tool for applications of various programming languages and of different programming paradigms. The authors describe a method for the horizontal integration of KBMS applications from different programming languages. They used the commonly known ADT definition technique to maintain a vertical integration between languages and the KBMS, and propose schema sharing and polymorphism as the background for this particular kind of interoperability. The suggested approach was validated by an extended KBMS object manager, the conceptual object manager. Some of the design principles of the conceptual object manager, the vertical integration task, and the semantics of the horizontal integration are considered.>

A connection between a world model about typing concepts of programming languages and an application domain model is outlined in order to conveniently generate a conceptual knowledge-base management system (KBMS) scheme for AI applications. The world model is defined in terms of a knowledge representation language and shall be incorporated into an interoperability supporting KBMS object manager. The application domain model is modeled in the same knowledge representation language. Their linkage results in a convenient generation of a conceptual KBMS schema for applications of the modeled domain. These applications can be written in different application languages and are allowed to share their individual KBMS schemata generated from the conceptual one.>