Research On The Method Of Multi-source Network Toponyms Dynamic Integration

ABSTRACT:Toponymy is an important part of the Geographic Information Public Service Platform in "Digital Peripherals". One hand the social development brought great challenges to the construction and maintenance for the traditional geographic name databases, on the other hand thedevelopment of technologg, especially the volunteer geographic information (VGI) technology, provides a new way for toponymy information collection. Geographic name databases is a true reflection of the real world in a regional features, the contents don’t include GIS framework features only, but also include non-frame attributes. Traditional network geographic databases come from mapping departments, geographic information is collected in accordance with the framework. In Web2.0VGI databases is from the general public contents, with a large number of non-geographic-framework information. Therefore, the integration of net-work geographic databases and VGI databases can generate content-rich geographic name databases, and updating the existing geographic name databases by VGI data can make up traditional geographic framework and keep the data up to date. But many differences in logic, semantic system, format with geographic data between the network data sources and VGI has hindered the integration of multi-source geographic name databases. Based on this consideration, the author studied and implemented the integration and updating for multi-source network name databases and design an algorithm to differentiate the data in Visualization. The main work contribution of this thesis are listed as follows:(1) The comparison between the traditional network name databases (GeoNames, GNS) and VGI from the following aspects, e.g., content richness, data accuracy, and update cycle. The GNS, GeoNames, OSM three network names data source are chosen to be the surrounding geographical names data source overseas, GeoNames, OSM data are chosen to be the update-incremental sources of name databases.(2) Against the problem for low degree of automation and high overlapping in the current geographical names labeling, propose a new automatic labeling method taking the weight of the place names into account, which establishes the Weight-Knowledge Base on the classification of geographical names, and then finds the best elements in the Weight-Knowledge Base according to the attribute of layer’s classification, and groups all labels of geographical names based on the weight of knowledge, and carries on overlay analysis and optimization analysis based on the weight of each group of labels when labeling, and thus achieve optimal labels.(3) A prototype system based on GDAL of integration and updating on multi-source network toponymys is implemented with Visual Studio2010, C#and Windows platform. By achieving data integration for network toponymys, update, and names labeling algorithm taking into account the weights, it verifies the correctness and feasibility of the method.