Approaches to the integration of digital geological information, obtained from the different sources
|Location||International Geological Congress,oslo 2008|
|Author||Florenskiy, Kirill۱; Egorova, Irina۲; Stavskiy, Analoly۲|
|Holding Date||11 October 2008|
In the process of development and operating of the information systems, whose subject area concerns geology, very frequently appears the need for the integration of digital geological data from different sources.
It is possible to designate such typical tasks of integration as the agreement of metamodels, the interpretation of the data models (ontologies), the comparison of classifiers, dictionaries and domains and the association of factographic information.
As an example it is possible to give the Electronic Chart of Subsoil Use of Russia, developed by "Mineral" Center in 2006 as the tool of complex data analysis from the main federal information arrays, such as the State Balance of Mineral Resources of the Russian Federation, State Cadastre of Mineral Deposits of Russian Federation, the State Registry of Subsoil Use Licenses and the Electronic Map of Licensed Areas for Oil and Gas Exploration.
Difficulties of the integration
Agreement of metamodels does not cause the serious difficulties since either the relational or object-oriented approaches, which are sufficiently well compatible, are used. However solution of remaining tasks often causes such problems, as conceptual differences in the data models (ontologies), used in the different data sources in the limits of one subject area, partially or completely overlapped, but badly compatible classifiers, dictionaries and domains.
Typical example is the difference in the interpretation of term "deposit" in the State Balance of Mineral Resources and in the State Cadastre of Mineral Deposits of Russian Federation. Object that Cadastre calls "deposit" (giving it the corresponding identifying number) is often called "group of deposits" in the Balance.
Choosing the best approach to solution
Procedural method assumes the hard coding of the procedures of integration, based on the given data models (ontologies). When it is expedient to use: if the tasks of interpretation are irregular, not predicted and repeated use is not planned. We have chosen this way for the rapid initial integration of data from the sources in the first version of the Electronic Chart of Subsoil Use of Russia.
Declarative method correlates the data models (ontologies) between themselves at the level of the description, processed by the standard procedure, which is independent of the data models. When it is expedient to use: when it’s possible to consider the need for the integration in the design stage of data sources.
Combined procedural- declarative method uses a procedural approach for the interpretation of data of each of the sources to transfer it into universal metamodel of higher level, in which the data models (ontologies) are described uniformly. When it is expedient to use: when the tasks of integration are sufficiently predictable and regular, i.e. regular data refreshment of the given information system. This approach is now used by us for the regular refreshment of the Electronic Chart of Subsoil Use of Russia.