CONSTRUCTION OF GIS-INTEGRATED SYSTEM OF DATA AND MODELS, BASED ON XML FORMALIZATION, FOR SIMULATING PROCESSES, TAKING PLACE IN RIVERS
Keywords:geoinformation system, mathematical model, database, XML, PMML, MathML, SBML, GML, WaterML, KNIME, simulation of the processes, river
The paper presents the conducted analysis of the known formats for formalization of mathematical models and spatial formats which are relevant for processing data and models occurring in rivers, namely, PMML, MathML, SBML, GML, WaterML, according to the system of the following criteria:availability of repositories of the already identified models, integration with programming languages, availability of specific tags for working with spatial data, availability of environments for automated formalization and import/export of XML models. An integral criterion is proposed. Examples of using this criteria system for selection of the optimal XML format for storing data and models are presented for different weights of certain criteria and conditions: if availability of geo-reference is the main criterion, Water ML will the optimal XML language, if it is the possibility to formalize any analytical models, especially hydro-biological ones, then SBML will be the optimal choice and if automation of artificial intelligence algorithms is required, then PMML should be chosen.
The following known technologies for storing attributive and spatial data of GIS for their automated processing are characterized: KML, Shapefile, GPX, GeoJSON, SXF, ArcGIS geo-data base, Spatialite (SQLite), MapInfo TAB format. It is noted that by the criterion of the number of references with the word “map” found by Google search system such formats as KML (22 million), GPX (20 million) and Shapefile (11,5 million) are the most popular in the world.
GIS integrated system of data and models based on XML formalization is proposed for the first time and its operability is illustrated by the example of predicting average annual water consumption over a multiyear period for 50 % provision in the Dniester river basin in KNIME Analytics Platform environment.
The obtained results make it possible to provide fast construction, versatility and broad functional of GIS-integrated system of data and models.