A. Motro. Multiplex: A Formal Model for Multidatabases and Its Implementation. In International Workshop on Next Generation Information Technology and Systems, pages 138--158. Springer-Verlag, LNCS 1649, 1999.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....kinds of integrity constraints. Imielinski et al. 63] provide a comprehensive characterization of the computational complexity of evaluating conjunctive queries in databases with OR objects. Those results carry over into our framework only in very limited cases, as discussed in [7,28] Motro [83] addressed the issue of integrating data from possibly mutually inconsistent sources in a fashion di#erent from the above and closer to our approach. He proposed, among others, the notion of sound query answers the answers present in the query result in every source. For functional dependencies ....

....not true, since a tuple that appears only in a single source will not be a sound answer, while it is a consistent answer if it does not conflict with any other tuple. Also, for general denial constraints, there may be sound answers that are not consistent. The computational mechanism proposed in [83] consists of simply taking the intersection of the query answers in individual sources and thus it is local. No complexity analysis is provided. Gertz [49,50] described techniques based on model based diagnosis for detecting causes of inconsistencies in databases and computing the corresponding ....

[Article contains additional citation context not shown here]

A. Motro. Multiplex: A Formal Model for Multidatabases and Its Implementation. In International Workshop on Next Generation Information Technology and Systems, pages 138--158. Springer-Verlag, LNCS 1649, 1999.

....the integrated one in toto, without any changes. There is no need for introducing disjunctive information. The integrity constraints are used only during querying. Second, the above approaches do not generalize to arbitrary functional dependencies and other kinds of integrity constraints. Motro [66] addressed the issue of integrating data from possible inconsistent sources in a fashion di#erent from the above and closer to our approach. He proposed, among others, the notion of sound query answers the answers present in the query result in every source. For functional dependencies and ....

Motro, A. Multiplex: A Formal Model for Multidatabases and Its Implementation. In International Workshop on Next Generation Information Technology and Systems, pages 138--158. Springer-Verlag, LNCS 1649, 1999.

....multiple heterogeneous distributed sources [2 8] 2. 2.2. 2. Inform when new data of special interest becomes available [9] 3. 3.3. 3. Negotiate for, purchase and receive information, goods and services [10 14] 4. 4.4. 4. Explain the relevance, quality and reliability of that information [15, 16], and 5. 5.5. 5. Adapt, learn and evolve in response to changing conditions [17 25] What is needed is an architecture that incorporates the concepts of infomediation with the flexibility of agents. In this paper, we present an architecture to create advanced information systems that offers an ....

....resources one would expect that sources would overlap in providing similar but inconsistent data. Inconsistencies are detected during the process of integration, and harmonization agents are engaged to resolve them. The concept of the harmonization agent is incorporated in the Multiplex [16, 52] proof of concept system which considers the reliability and quality of the conflicting information sources, and resolves conflicts in a way that increases the overall value of the information [15, 53 55] 6 3 Thesaurus Services for Infomediation in the InHead Architecture In this section we ....

A. Motro, Multiplex: A Formal Model for Multidatabases and Its Implementation, ISSE Department, George Mason University, Fairfax, VA, Technical Report ISSE-TR-95-10, 1995.

....the research literature for approximate query answering in situations where answers are obtained from multiple information sources. Most approaches attempt to estimate some measure of divergence from the true answer and are based on some technique of modeling uncertainty. In the Multiplex project [20], the soundness and completeness of the results are estimated based on the intersections and unions of the candidate results. In our approach, the Information Retrieval analogs of soundness (precision) and completeness (recall) are estimated based on the sizes of the extensions of the terms. We ....

A. Motro. Multiplex: A Formal Model of Multidatabases and its Implementations. Technical report, Technical Report ISSE-TR-95-103, Department of Information and Software Systems Engineering, George Mason University, Fairfax, Virginia, March 1995.

....approximating query answering in situations where multiple answers may be obtained from multiple information sources. Most approaches are typically accompanied by an attempt to estimate some measure of divergence from the true answer and are based on modeling uncertainty. In the Multiplex project [30], the soundness and completeness of the results are estimated based on the intersections and unions of the candidate results. In our approach, the information retrieval analogs of soundness (precision) and completeness (recall) are estimated based on the sizes of the extensions of the terms. We ....

A. Motro. Multiplex: A formal model of multidatabases and its implementations. Technical report, Technical Report ISSE-TR-95-103, Department of Information and Software Systems Engineering, George Mason University, Fairfax, Virginia, March 1995.

....answering in situations where multiple answers may be obtained from multiple information sources. Most approaches are typically accompanied by an attempt to estimate some measure of divergence from the true answer and are based on some technique of modeling uncertainty. In the Multiplex project [16], the soundness and completeness of the results are estimated based on the intersections and unions of the candidate results. In our approach, the Information Retrieval analogs of soundness (precision) and completeness (recall) are estimated based on the sizes of the extensions of the terms. We ....

A. Motro. Multiplex: A formal model of multidatabases and its implementations. Technical report, Technical Report ISSE-TR-95-103, Department of Information and Software Systems Engineering, George Mason University, Fairfax, Virginia, March 1995.

....to deal with instance level conflicts but it requires the applications to use a non standard data model for data access. This approach only deals with conflicts at predicate evaluation time and the tolerance mode is always HighConfidence. Conflicts in query results are not removed. Multiplex [5] deals with instance level conflicts in the context of answering queries using given materialized views. Conflicts arise when the materialized views overlap and the same query can be evaluated in multiple ways resulting in multiple answers. c1 = HighConfidence c1 = RandomEvidence c1 = ....

A. Motro. Multiplex: A formal model for multidatabases and its implementation. Technical Report ISSE-TR-95-103, Department of Information and Software System Engineering, George Mason University, 1995.

....increase the availability of all data sources to the point that queries almost always execute. However, note that parachute queries are completely compatible with replication in the case that a data source is replicated, the probability that it will be unavailable is simply smaller. Multiplex [17] tackles the issue of unavailable data sources in a multidatabase system and APPROXIMATE [21] tackles the issue of unavailable data in a distributed database. Both systems propose an approach based on approximate query processing. In presence of unavailable data, the system returns an approximate ....

A. Motro. Multiplex: A formal model for multidatabases and its implementation. Technical Report ISSE-TR-95-103, George Mason University, 1995.

....answering in situations where multiple answers may be obtained from multiple information sources. Most approaches are typically accompanied by an attempt to estimate some measure of divergence from the true answer and are based on some technique of modeling uncertainty. In the Multiplex project [16], the soundness and completeness of the results are estimated based on the intersections and unions of the candidate results. In our approach, the Information Retrieval analogs of soundness (precision) and completeness (recall) are estimated based on the sizes of the extensions of the terms. We ....

A. Motro. Multiplex: A formal model of multidatabases and its implementations. Technical report, Technical Report ISSE-TR-95-103, Department of Information and Software Systems Engineering, George Mason University, Fairfax, Virginia, March 1995.

....has been done in the area of intergation of databases[10] missing from this research has been a uniform framework; each approach formulates its own version of the problem, states its own assumptions, develops its own ad hoc model, and proposes its own solutions. The Multiplex multidatabase model[6] addresses many of these issues and will be the model of choice for our research. In this paper we assume that all intensional inconsistencies are resolved in the framework of the Multiplex multidatabase model, and that the databases participating in the integrated multidatabase are accessed via a ....

....selections use only range conditions, and projections always retain the key attribute(s) 2) The stored information (the database instance) is relatively static, and hence the quality of data does not change frequently. 2. 2 Reconciling Inconsistent Answers in Multidatabases We employ Multiplex [6] as our formal model of multidatabases. Multiplex is an extension of the relational model. It extends the definition of schemes, constraints, queries, and answers to an environment of multiple databases, and it provides a common ground for other definitions as they become necessary. Data quality ....

A. Motro. Multiplex: A formal model for multidatabases and its implementation. Technical report, Dept. of Information and Software Systems Engineering, George Mason University, March 1995.

.... finite representations of infinite sets of databases, both in the constraint database literature [KKR90,vdM93] and in the literature on uncertain and indefinite databases [AKG91,Gra91,IL84,Men84] We will not go into details; for surveys of these fields see [Kan95] and [vdM98] The work of Motro [Mot97] is close in spirit and motivation to ours; however, Motro assumes the existence of a real world global database instead of modeling the uncertainty in the sources by associating a set of databases with them; this makes it difficult to give precise meaning to his ideas. In a recent paper [AD98] ....

.... instance window function [MRW86] with no dependencies, and if all sources are clopen, then the only possible global database is the notorious universal relation [Ull82] Acknowledgments We thank NSERC, IRIS and Concordia FRDP for their support, and Anthony Tomasic for pointing out Motro s work [Mot97]. ....

Motro A. Multiplex: A Formal Model for Multidatabases and Its Implementation. Technical Report ISSE-TR-95-103.

....set fAE C gC evaluates truthness of the theories, c.f. Ginsberg 1988) Partial contexts and worlds are related to contexts of Formal Concept Analysis (FCA) see (Ganter Wille 1996) The rough sets theory (Pawlak 1991) also emphasises partial, or boundary objects. In the area of databases, (Motro 1995) has a similar focus. ....

Motro, A. 1995. Multiplex: A formal model for multidatabases and its implementation. Technical report, Dept. of Inf. and Soft. Syst. Eng., George Mason Univ.

....sources, replication may not be possible simply because the data source forbids it. However, note that incremental query processing is completely compatible with replication in the case that a data source is replicated, the probability that it will be unavailable is simply smaller. Multiplex [13] tackles the issue of unavailable data sources in a multi database system and APPROXIMATE [16] tackles the issue of unavailable data in a distributed database. Both systems propose an approach based on approximate query processing. In presence of unavailable data, the system returns an approximate ....

A. Motro. Multiplex: A formal model for multidatabases and its implementation. Technical Report ISSE-TR-95-103, George Mason University, 1995.

....as one of the answers to the above LOREL query. 7. Related Work Various projects on the integration of heterogeneous sources (see Ahmed et al. 1991] Kim et al. 1993] Batini et al. 1986] Hammer et al. 1993] Litwin et al. 1990] Thomas et al. 1990] Gupta [1989] Su et al. 1996] and Motro [1995]) focus on integrating relatively small numbers of structured databases. Most of the research on these projects focuses on resolving the semantic and schematic heterogeneities that arise upon integration. The solutions often rely on semantically rich data models that allow for easy representation ....

A. Motro [1995]. "Multiplex: A Formal Model for Multidatabases and its Implementation," Technical Report ISSE-TR-95-103, George Mason University, 1995.

....of Section 6. Section 7 summarizes the findings and outlines ideas for further research. We begin with a brief discussion of virtual databases and the Autoplex discovery system for virtual databases. 2 Background Our work is conducted within the framework of the Multiplex virtual database system [10] and the Autoplex content discovery system [2] and in this section we provide brief descriptions of both. We note, however, that much of our work is of general applicability to most such database integration systems. 2.1 Multiplex The multidatabase system Multiplex [10] is an example of a ....

....virtual database system [10] and the Autoplex content discovery system [2] and in this section we provide brief descriptions of both. We note, however, that much of our work is of general applicability to most such database integration systems. 2. 1 Multiplex The multidatabase system Multiplex [10] is an example of a virtual database system. The basic architecture of Multiplex is fairly simple (Figure 1) Whereas a traditional database consists of schema and data that conforms to the schema, a Multiplex database consists of schema and mapping that describes data that is stored elsewhere. ....

[Article contains additional citation context not shown here]

Amihai Motro. Multiplex: A formal model for multidatabases and its implementation. In Proceedings of the Fourth International Workshop on Next Generation Information Technologies and Systems, pages 138--158, 1999.

....together. Under these new assumptions, a single, complete answer can no longer be guaranteed. It is now possible that a query could not be answered in its entirety, or it might result in several different answers. Multiplex is a multidatabase model designed to operate under these new assumptions [14]. The advantages of the Multiplex model are numerous: 1) The global scheme, and hence its user interfaces, are based on the popular relational model. 2) It supports heterogeneity; that is, the source databases may be structured according to any database model, as long as they can communicate ....

....of Multiplex is its handling of troublesome situations in which there is too little data (e.g. some sources are unavailable) or too much data (i.e. some sources are inconsistent) Implementation. At the present, the Multiplex system is being upgraded from an earlier version described in [14] to the version that was described here. We mention here some of the architectural features common to both versions. Multiplex extends the query language of conjunctive queries with aggregate functions. A language based on conjunctive queries with aggregation provides a fairly powerful querying ....

A. Motro. Multiplex: A formal model for multidatabases and its implementation. Technical Report ISSE-TR-95-103, Department of Information and Software Systems Engineering, George Mason University, March 1995.

.... autonomous and heterogeneous databases [5, 11] The usual focus in systems that address this problem is the resolution of intensional inconsistencies: How to reconcile the schematic differences among the participating databases (this issue is known as semantic heterogeneity) The Multiplex project [7] also addresses the problem of extensional inconsistencies: How to reconcile the data differences among the participating databases. Specifically, Multiplex does not assume that overlapping information sources are mutually consistent, and it focuses on methods for harmonizing inconsistent answers ....

Motro, A: Multiplex: A Formal Model for Multidatabases and Its Implementation. Technical Report ISSE-TR-95-103, Department of Information and Software Engineering, George Mason University, March 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