Nikolay A. Skvortsov

Information Sources Registration at a Subject Mediator as Compositional Development

Method for heterogeneous information source registration at subject mediators with local as view (LAV) organization is presented. LAV approach considers schemas exported by sources as materialized views over virtual classes of the mediator. This approach is intended to cope with a dynamic, possibly incomplete set of sources. To disseminate the information sources, their providers should register them at a respective subject mediator. Such registration can be done concurrently and at any time. The registration method proposed is new and contributes to the following. The method is applicable to wide class of source specification models representable in hybrid semistructured/object canonical mediator model. Ontological specifications are used for identification of mediator classes semantically relevant to a source class. Maximal subset of source information relevant to the mediator classes is identified. Concretizing types are defined so that federated classes instance types are refined by the source instance type. This direction naturally supports query planning refining a mediator query in terms of a specific source. Such refining direction is in contrast to conventional compositional development where specification of requirements is to be refined by specifications of components. Such inversion is natural for the registration process: a materialized view (requirements) is constructed over virtual specifications (components).

Conceptual approach to astronomical problems

New technical capabilities have brought about the sweeping growth of the amount of data acquired by the astronomers from observations with different instruments in various parts of the electromagnetic spectrum. We consider conceptual approach to be a promising tool to efficiently deal with these data. It uses problem domain knowledge to formulate the tasks and develop problem-solving algorithms and data analysis methods in terms of domain concepts without reference to particular data sources, and thereby allows solving certain problems in general form. We demonstrate the benefits of conceptual approach by using it to solve problems related to search for secondary photometric standard candidates, determination of galaxy redshifts, creation of a binary and multiple star repository based on inhomogeneous databases, and classification of eclipsing binaries.We formulate and solve these problems over specifications of astronomical knowledge units such as photometric systems, astronomical objects, multiple stars, etc., and define them in terms of the corresponding problem domains independently of the existing data resources.

Designing personalized digital libraries over websites with semistructured data

Issues concerning the design of personalized digital libraries over collections of semistructured data available on the Web are considered. The approach suggested makes it possible to design libraries adjusted to the personal needs of information users that place various requirements upon the contents and representation of information. The digital library is designed as a composition of fragments of websites. In this paper, a method for the compositional design of information systems is applied to semistructured data on the Web. This method was developed at the Institute for Problems of Informatics of the Russian Academy of Sciences (IPI RAS) [2]. The design procedure is demonstrated by designing a library over two websites containing data about registered patents.

Matching and Verification of Multiple Stellar Systems in the Identification List of Binaries

Binary and multiple stellar systems have been observed using various methods and tools. Catalogs of binaries of different observational types are independent and use inherent star identification systems. Catalog rows describing components of stellar systems refer to identifiers of surveys and catalogs of single stars. The problem of cross-identification of stellar objects contained in sky surveys and catalogs of binaries of different observational types requires not only combining lists of existing identifiers of binary stars, but rather matching components and of multiple systems and pairs of components by their astrometric and astrophysical parameters. Existing identifiers are verified for belonging to matched components, pairs and systems. After that, they may be matched to one another. The framework of multiple system cross-matching presented in the paper uses domain knowledge of binaries of different observational types to form sets of matching criteria. The Identification List of Binaries (ILB) has been created after accurate matching of systems, their components and pairs of all observational types. This work continues research of binary and multiple system identification methods.

Binary Star Database (BDB): New Developments and Applications

Binary star DataBase (BDB) is the database of binary/multiple systems of various observational types. BDB contains data on physical and positional parameters of 260,000 components of 120,000 stellar systems of multiplicity 2 to more than 20, taken from a large variety of published catalogues and databases. We describe the new features in organization of the database, integration of new catalogues and implementation of new possibilities available to users. The development of the BDB index-catalogue, Identification List of Binaries (ILB), is discussed. This star catalogue provides cross-referencing between most popular catalogues of binary stars.

Conceptualization of Methods and Experiments in Data Intensive Research Domains

Nowadays research of various scopes especially in natural sciences requires manipulation of big volumes of data generated by observation, experiments and modeling. Organization of data-intensive research assumes definition of domain specifications including concepts (specified by ontologies) and formal representation of data describing domain objects and their behavior (using conceptual schemes), shared and maintained by communities working in the respective domains. Research infrastructures are based on domain specifications and provide methods applied to such specifications, collected and developed by research communities. Tools for organizing experiments in research infrastructures are also supported by conceptual specifications of measuring and investigating object properties, applying the research methods, describing and testing the hypotheses. Astronomy as a sample data intensive domain is chosen to demonstrate building of conceptual specifications and usage of them for data analysis.

Value Reconciliation in Mediators of Heterogeneous Information Collections Applying Well-Structured Context Specifications

Method for value reconciliation in Local as Views (LAV) mediators of heterogeneous information collections applying well-structured specifications is presented. This approach extends a procedure for heterogeneous information sources registration at subject mediators with LAV organization. Conflicts in value semantics and representation in contexts of the mediator and a collection should be recognized and explicitly specified. According to the proposed method, value semantics contexts should be defined for the mediator and for a collection. In such context for each value kind a type is to be specified that includes a generic function converting values from a collection context into the mediator context (and/or back). Each type attribute having the respective value semantics is to be typed with such type definition. Such structuring of type attribute semantics definition gives significant economy in development of value conversion functions needed to reconcile values between the mediator and collection contexts.