Intelligent Search of Spatial Data Analysis Context

摘要

The paper deals with the task of controlling the course of interactive analysis of spatial data using geographic information systems. The main problem of analysis is considered a significant amount of data that the analyst receives on requests to the information database of the geographic information system. The goal of the management is to provide the user with a set of cartographic images that are as useful as possible for the analysis. The concept of semantic content of images and analysis contexts is introduced. A conceptual knowledge model of the analysis context is proposed. Its distinguishing feature is the presence of two components. The first component is the core of the context. It includes fundamental knowledge of the method of analysis. Knowledge determines the spatial, temporal and semantic boundaries of the field of analysis. The second component is valid context conversions that preserve its meaning. A condition is introduced to preserve the meaning of the analysis process as the correspondence of the requested data to the limit of permissible transformations. An algorithm for searching the context that preserves the semantic content of the analysis is given. The algorithm uses an archive of use cases of sequences of change of contexts. The results of an experimental study of the effectiveness of the method are given on the example of a corporate geographic information system.